Entries |
Document | Title | Date |
20080205750 | Method for Adaptively Boosting Classifiers for Object Tracking - A method adapts a boosted classifier to new samples. A boosted classifier is trained using initial samples. The boosted classifier is a combination of weak classifiers. Each weak classifier of the boosted classifier is updated adaptively by adding contributions of new samples and deleting contributions old samples. | 08-28-2008 |
20080232681 | OBJECT DETECTION SYSTEM BASED ON A POOL OF ADAPTIVE FEATURES - A method, system and computer program product for detecting presence of an object in an image are disclosed. According to an embodiment, a method for detecting a presence of an object in an image comprises: receiving multiple training image samples; determining a set of adaptive features for each training image sample, the set of adaptive features matching the local structure of each training image sample; integrating the sets of adaptive features of the multiple training image samples to generate an adaptive feature pool; determining a general feature based on the adaptive feature pool; and examining the image using a classifier determined based on the general feature to detect the presence of the object. | 09-25-2008 |
20080232682 | SYSTEM AND METHOD FOR IDENTIFYING PATTERNS - The present invention provides a system and method for identifying a pattern as belonging to one of a set of predetermined classes of patterns. The system comprises a plurality of classifier blocks wherein each classifier block corresponds to a distinct predetermined class of patterns and produces a mirror image of an input pattern if the input pattern belongs to the predetermined class. The system also comprises a plurality of sub-classifier blocks wherein each sub-classifier block corresponds to a distinct predetermined sub-class of a predetermined class of patterns and is coupled to a classifier block corresponding thereto for producing a mirror image of an input pattern if the pattern belongs to the predetermined sub-class. The system further comprises an input unit for capturing the pattern for identification and an output unit for displaying at least one of a mirror image of an input pattern and an identified class and sub-class of the input pattern. The system and method of the present invention may also be used as sub-modules for building large generalized learning systems. | 09-25-2008 |
20080240551 | LOCAL BI-GRAM MODEL FOR OBJECT RECOGNITION - A local bi-gram model object recognition system and method for constructing a local bi-gram model and using the model to recognize objects in a query image. In a learning phase, the local bi-gram model is constructed that represents objects found in a set of training images. The local bi-gram model is a local spatial model that only models the relationship of neighboring features without any knowledge of their global context. Object recognition is performed by finding a set of matching primitives in the query image. A tree structure of matching primitives is generated and a search is performed to find a tree structure of matching primitives that obeys the local bi-gram model. The local bi-gram model can be found using unsupervised learning. The system and method also can be used to recognize objects unsupervised that are undergoing non-rigid transformations for both object instance recognition and category recognition. | 10-02-2008 |
20080247639 | Data Matching Method, Data Matching Apparatus and Data Matching Program - The purpose is to execute the matching of a data after state change from a data before state change, or the matching of a data before state change from a data after state change. The component analyzing unit realized by software analyzes an input data by using a configuration component of a selected state-specific database and sends the configuration component coefficient to the component coefficient conversion unit. The component coefficient conversion unit converts the sent configuration component coefficient into the configuration component coefficient corresponding to the state-specific database which is not selected and sends it to the state change data generation unit. The state change data generation unit generates the state change data by using the sent configuration component coefficient and the configuration component of the corresponding state-specific database. The matching unit matches the state change data and the registration data accumulated in the matching data accumulating unit. | 10-09-2008 |
20080253645 | Adaptive Classifier, and Method of Creation of Classification Parameters Therefor - A method of generating classifier parameters from a plurality of multivariate sample data, for use in subsequent classification, said classifier parameters relating to a plurality of intervals on each of the variables, said intervals being associated with classes, comprising: inputting said sample data; calculating a plurality of boundaries for each of said variables from said sample data, and deriving parameters defining said intervals from said boundaries. | 10-16-2008 |
20080260240 | User interface for inputting two-dimensional structure for recognition - In embodiments consistent with the subject matter of this disclosure, a user may input one or more strokes as digital ink to a processing device. The processing device may produce and present a recognition result, which may include a misrecognized portion. A user may indicate a desire to correct the misrecognized portion and may further select one or more strokes of the misrecognized portion. The processing device may then present the one or more recognition alternates corresponding to the selected one or more strokes of the misrecognized portion. In some embodiments, the processing device may permit a user to rewrite the selected one or more strokes of the misrecognized portion with newly entered digital ink. Features, such as, rewriting and correction of the input digital ink may be discoverable in some embodiments. | 10-23-2008 |
20080260241 | Grouping writing regions of digital ink - A method for grouping writing regions of digital ink receives processed digital ink that comprises writing regions. A relationship can be generated between a plurality of the writing regions. A feature set can be determined that is associated with the plurality of the writing regions. The plurality of the writing regions can be grouped based on the feature set. | 10-23-2008 |
20080292181 | Information Processing Method, Information Processing Apparatus, and Storage Medium Storing a Program - An information processing method includes: for image data of each of a plurality of images, obtaining scene information concerning the image data from supplemental data that is appended to the image data, classifying a scene of an image represented by the image data, based on the image data, comparing the classified scene with a scene indicated by the scene information; and if there is a mismatch image for which the classified scene does not match the scene indicated by the scene information, displaying information concerning the mismatch image on a confirmation screen. | 11-27-2008 |
20080317335 | METHOD OF IDENTIFICATION ACCORDING TO SELECTED PATTERNS AND RELATED COMPUTER SYSTEM - A method of identification of a computer system includes: (a) selecting a first partial pattern from a pattern group; (b) determining whether the first partial pattern selected in step (a) corresponds with a predetermined rule; and (c) controlling the computer system whether to operate a program according to a determining result in step (b). | 12-25-2008 |
20090016599 | SEMANTIC REPRESENTATION MODULE OF A MACHINE-LEARNING ENGINE IN A VIDEO ANALYSIS SYSTEM - A machine-learning engine is disclosed that is configured to recognize and learn behaviors, as well as to identify and distinguish between normal and abnormal behavior within a scene, by analyzing movements and/or activities (or absence of such) over time. The machine-learning engine may be configured to evaluate a sequence of primitive events and associated kinematic data generated for an object depicted in a sequence of video frames and a related vector representation. The vector representation is generated from a primitive event symbol stream and a phase space symbol stream, and the streams describe actions of the objects depicted in the sequence of video frames. | 01-15-2009 |
20090016600 | COGNITIVE MODEL FOR A MACHINE-LEARNING ENGINE IN A VIDEO ANALYSIS SYSTEM - A machine-learning engine is disclosed that is configured to recognize and learn behaviors, as well as to identify and distinguish between normal and abnormal behavior within a scene, by analyzing movements and/or activities (or absence of such) over time. The machine-learning engine may be configured to evaluate a sequence of primitive events and associated kinematic data generated for an object depicted in a sequence of video frames and a related vector representation. The vector representation is generated from a primitive event symbol stream and a phase space symbol stream, and the streams describe actions of the objects depicted in the sequence of video frames. | 01-15-2009 |
20090041340 | Image Processing System, Learning Device and Method, and Program - The present invention relates to an image processing system, a learning device and method, and a program which enable easy extraction of feature amounts to be used in a recognition process. Feature points are extracted from a learning-use model image, feature amounts are extracted based on the feature points, and the feature amounts are registered in a learning-use model dictionary registration section | 02-12-2009 |
20090052768 | IDENTIFYING A SET OF IMAGE CHARACTERISTICS FOR ASSESSING SIMILARITY OF IMAGES - The invention relates to a method ( | 02-26-2009 |
20090074288 | DATA PROCESSING APPARATUS, COMPUTER PROGRAM PRODUCT, AND DATA PROCESSING METHOD - A data processing apparatus includes a feature-value calculating unit that calculates an image feature value indicating a feature of image data, a case database including a case set including a correspondence of image feature values and functions, and an optimum-function predicting unit that predicts an optimum function based on the case database and the image feature value calculated by the feature-value calculating unit. Due to the optimum-function predicting unit, work efficiency of a user can be improved. | 03-19-2009 |
20090087084 | METHOD AND APPARATUS FOR PATTERN RECOGNITION - Methods and apparatuses for pattern recognition involve quantum-mechanical calculations. Pattern recognition can be achieved by considering a quantum system and its Hamiltonian dynamics. The dynamics are calculated on the basis of an initial Hamiltonian indicating an initial quantum state and on the basis of a final Hamiltonian. The final Hamiltonian depends on an input pattern and reference patterns. Transformations according to the Hamiltonian dynamics for the quantum system are applied to generate a final quantum state of said quantum system. Depending on said final quantum state a similarity between said input pattern and said reference patterns is determined. | 04-02-2009 |
20090087085 | TRACKER COMPONENT FOR BEHAVIORAL RECOGNITION SYSTEM - A tracker component for a computer vision engine of a machine-learning based behavior-recognition system is disclosed. The behavior-recognition system may be configured to learn, identify, and recognize patterns of behavior by observing a video stream (i.e., a sequence of individual video frames). The tracker component may be configured to track objects depicted in the sequence of video frames and to generate, search, match, and update computational models of such objects. | 04-02-2009 |
20090092312 | Identifying Method and Storage Medium Having Program Stored Thereon - An identification processing that matches a preference of a user is carried out. An identifying method according to the present invention is an identifying method, in which learning is carried out using a learning sample and, based on a learning result, identification is performed as to whether or not a target of identification belongs to a certain class, including: extracting a learning sample belonging to the certain class and a learning sample not belonging to the certain class, displaying a plurality of the extracted learning samples arranged on a display section, as well as displaying a mark between the learning sample belonging to the certain class and the learning sample not belonging to the certain class, and displaying the mark between a different pair of the learning samples by moving a position of the mark in response to an instruction of a user, changing an attribute information that indicates a class to which the learning sample belongs in response to the position of the mark determined by the user, and identifying whether or not a target of identification belongs to the certain class based on a result of relearning using the learning sample of which the attribute information has been changed. | 04-09-2009 |
20090097739 | PEOPLE DETECTION IN VIDEO AND IMAGE DATA - A process identifies a person in image data. The process first executes a training phase, and thereafter a detection phase. The training phase learns body parts using body part detectors, generates classifiers, and determines a spatial distribution and a set of probabilities. The execution phase applies the body part detector to an image, combines output of several body part detectors, and determines maxima of the combination of the output. | 04-16-2009 |
20090097740 | Method and Apparatus for Automating an Inspection Procedure - A process for using a hand-held infrared inspection system incorporating on-board training, on-board validation, on-board operator certification, on-board reporting information, or on-board survey instructions. Improved methods for automating area surveys are provided through exception-driven surveillance practices. Imbedded information enables less experienced operators to use more sophisticated devices more effectively. Validation or certification assures operator knowledge or ability. Multilevel classification of anomalies aids in automated analysis and report generation. | 04-16-2009 |
20090097741 | SMOTE ALGORITHM WITH LOCALLY LINEAR EMBEDDING - A data classification method. The method includes: providing data mapped in a first space; mapping the data into a second space using locally linear embedding to generate mapped data; applying a synthetic minority over-sampling technique (SMOTE) to the mapped data to generate new data; and mapping the new data into the first space. | 04-16-2009 |
20090110268 | TABLE OF CONTENTS EXTRACTION BASED ON TEXTUAL SIMILARITY AND FORMAL ASPECTS - An initial organizational table for a document is determined based on textual similarity between entries of the organizational table and target text fragments and not taking into account text formatting. A classifier is trained to identify text fragment pairs consisting of entries of the organizational table and corresponding target text fragments based at least in part on text formatting features. The training employs a training set of examples annotated based on the initial organizational table. The initial organizational table is updated using the trained classifier. | 04-30-2009 |
20090116734 | IMAGE CLASSIFICATION - An apparatus and method are provided for classifying elements in an image, in particular elements of a hyperspectral image, where an element is defined by a vector of feature values. The apparatus includes a classifier arrangement comprising a number of classifiers each operable, in respect of an element to be classified, to receive a different predetermined subset of the feature values from the element feature vector and wherein, in operation, each classifier is trained in respect of a predetermined set of classes using training data representative of elements in each class; and a combining arrangement operable to combine outputs from the classifiers to determine which of the predetermined classes to associate with an element to be classified, wherein each of the different predetermined subsets of feature values comprise a different cyclic selection of the feature values such that, in operation, adjacent feature values in an element feature vector are input to different ones of the classifiers and all feature values are input to at least one classifier. | 05-07-2009 |
20090116735 | WARNING APPARATUS AND METHOD FOR AVOIDING EYE STRESS - An exemplary warning method for avoiding eye stress of a computer user includes: capturing a number of consecutive images of the face of a computer user; processing the images to obtain a number of values each indicative of a degree of openness of the eyes of the computer user; counting an amount of values exceeding a predetermined threshold to obtain accumulated viewing time of the computer user; and triggering a warning means if the accumulated viewing time of the computer satisfies a predetermined condition. | 05-07-2009 |
20090116736 | SYSTEMS AND METHODS TO AUTOMATICALLY CLASSIFY ELECTRONIC DOCUMENTS USING EXTRACTED IMAGE AND TEXT FEATURES AND USING A MACHINE LEARNING SUBSYSTEM - A document analysis system that automatically classifies documents by recognizing in each document distinctive features comprises a document acquisition system, a document recognition training system, a document classification system, a document recognition system, and a job organization system. The document acquisition system receives jobs wherein each job containing at least one electronic document. The document feature recognition system automatically extracts image and text features from each received document. The document classification system automatically classifies recognized electronic documents by finding the best match between the extracted features of each of the document and feature sets associated with each category of document. The document recognition training system automatically trains the feature set for each corresponding category of documents, wherein the training system using extracted features of unrecognized documents automatically modifies the feature set for a document category. The job organization system automatically organizes each job according to the document categories it contains. | 05-07-2009 |
20090123062 | Information processing apparatus, information processing method, and program - Disclosed herein is an information processing apparatus configured to classify time-series input data into N classes, including, a time-series feature quantity extracting section, N calculating sections, and a determination section. | 05-14-2009 |
20090154796 | SYSTEMS AND METHODS FOR HUMAN BODY POSE ESTIMATION - Systems and computer-implemented methods for use in body pose estimation are provided. Training data is obtained, where the training data includes observation vector data and corresponding pose vector data for a plurality of images. The observation vector data is representative of the images in observation space. The pose vector data is representative of the same images in pose space. Based on the training data, a model is computed that includes parameters of mapping from the observation space to latent space, parameters of mapping from the latent space to the pose space, and parameters of the latent space. The latent space has a lower dimensionality than the observation space and the pose space. | 06-18-2009 |
20090175531 | SYSTEM AND METHOD FOR FALSE POSITIVE REDUCTION IN COMPUTER-AIDED DETECTION (CAD) USING A SUPPORT VECTOR MACNINE (SVM) - A method for computer aided detection (CAD) and classification of regions of interest detected within HRCT medical image data includes post-processing machine learning to maximize specificity and sensitivity of the classification to realize a reduction in number of false positive detections reported. The method includes training a classifier on a set of medical image training data selected to include a number of true and false regions, wherein the true and false regions are identified by a CAD process, and automatically segmented, wherein the segmented training regions are reviewed by at least one specialist to classify each training region for its ground truth, i.e., true or false, essentially qualifying the automatic segmentation, wherein a feature pool is identified and extracted from each segmented region, and wherein the pool of features is processed by genetic algorithm to identify an optimal feature subset, which subset is used to train a support vector machine, detecting, within non-training medical image data, regions that are candidates for classification, segmenting the candidate regions, extracting a set of features from each segmented candidate regions and classifying the candidate region using the support vector machine after training in accordance with the optimal feature subset, and processing the set of candidate features. | 07-09-2009 |
20090175532 | Method and System for Creating Flexible Structure Descriptions - In one embodiment, the invention provides a method, comprising detecting data fields on a scanned image; generating a flexible document description based on the detected data fields, including creating a set of search elements for each data field, each search element having associated search criteria; and training the flexible document description using a search algorithm to detect the data fields on additional training images based on the set of search elements. | 07-09-2009 |
20090175533 | INFORMATION PROCESSING APPARATUS AND METHOD, RECORDING MEDIUM, AND PROGRAM - In an information processing apparatus, such as a robot that discriminates human faces, nodes are hierarchically arranged in a tree structure. Each of the nodes has a number of weak classifiers. Each terminal node learns face images associated with one label. An upper node learns learning samples of all labels learned by lower nodes. When a window image to be classified is input, discrimination is performed sequentially from upper nodes to lower nodes. When it is determined that the window image does not correspond to a human face, discrimination by lower nodes is not performed, and discrimination proceeds to sibling nodes. | 07-09-2009 |
20090196493 | Cognitive Memory And Auto-Associative Neural Network Based Search Engine For Computer And Network Located Images And Photographs - Designs for cognitive memory systems storing input data, images, or patterns, and retrieving it without knowledge of where stored when cognitive memory is prompted by query pattern that is related to sought stored pattern. Retrieval system of cognitive memory uses autoassociative neural networks and techniques for pre-processing query pattern to establish relationship between query pattern and sought stored pattern, to locate sought pattern, and to retrieve it and ancillary data. Cognitive memory, when connected to computer or information appliance introduces computational architecture that applies to systems and methods for navigation, location and recognition of objects in images, character recognition, facial recognition, medical analysis and diagnosis, video image analysis, and to photographic search engines that when prompted with a query photograph containing faces and objects will retrieve related photographs stored in computer or other information appliance, and will identify URL's of related photographs and documents stored on the World Wide Web. | 08-06-2009 |
20090208096 | TRANSFORMING MEASUREMENT DATA FOR CLASSIFICATION LEARNING | 08-20-2009 |
20090220146 | METHOD AND APPARATUS FOR CHARACTERIZING THE FORMATION OF PAPER - A method for characterizing the formation of paper in which patterns and/or structures existing in the paper are automatically characterized and classified. The automatic characterization and classification includes creating a collection of paper specimens, creating a digital image of each individual specimen, digital pre-processing of the digital image where necessary, calculating different multi-dimensional features in light of the digital images or sub-ranges of the images, analyzing structure-specific groups forming in the feature space during calculation of the different multi-dimensional features and analyzing the structure-specific groups in the feature space, projecting the results of the analysis of the structure-specific groups into a—compared to the feature space—low-dimensional space for visualizing the analysis results, and drawing on the analysis results for the classification of newly added specimens. The calculation of the different multi-dimensional features takes place in light of the digital images or sub-ranges of the images on the basis of at least one of the following algorithms: relational kernel function (RKF), phase-based method, 2-point or 3-point method, or wavelets. | 09-03-2009 |
20090232390 | Image processing device, image processing method, learning device, learning method, and program - An image processing device, to convert a first image data into a second image data having a higher image quality, includes: a predicted tap extracting unit to extract multiple pixels as a predicted tap for prediction computing; a level limit class tap extracting unit to extract multiple pixels as a level limit class tap for level limit classifying processing; a waveform class tap extracting unit to extract multiple pixels as a waveform class tap; a level limit classifying unit to classify the pixel of interest, based on a ratio between a level width and a dynamic range of the pixels making up the level limit class tap; a waveform pattern classifying unit to classify the pixel of interest; a prediction coefficient output unit to output a prediction coefficient corresponding to a combination of a level limit class and a waveform pattern class; and a prediction computing unit. | 09-17-2009 |
20090245625 | IMAGE TRIMMING DEVICE AND PROGRAM - An image trimming device involves: extracting a region of interest from an original image; detecting a set of features for each region of interest; determining whether each region of interest should be placed inside or outside a trimming frame based on the set of features and setting the trimming frame in the image; extracting an image inside the trimming frame; determining a positional relationship between each region of interest and the trimming frame and increasing or decreasing probability of each region of interest to be placed inside the trimming frame depending on if the region has a set of features similar to that of another region of interest previously placed inside the trimming frame or previously placed outside the trimming frame. | 10-01-2009 |
20090252405 | METHOD AND APPARATUS FOR DICTIONARY-BASED IMAGE PROCESSING - A method and system for processing a digitized image including multiple pixels is provided. Image processing involves determining a characteristic value for a set of image pixels, determining a classification for the set of pixels based on the characteristic value, accessing a dictionary including transforms, and retrieving transforms for the set of pixels based on said classification, and applying the transforms to the set of pixels to obtain output pixels. | 10-08-2009 |
20090263010 | ADAPTING A PARAMETERIZED CLASSIFIER TO AN ENVIRONMENT - A classifier is trained on a first set of examples, and the trained classifier is adapted to perform on a second set of examples. The classifier implements a parameterized labeling function. Initial training of the classifier optimizes the labeling function's parameters to minimize a cost function. The classifier and its parameters are provided to an environment in which it will operate, along with an approximation function that approximates the cost function using a compact representation of the first set of examples in place of the actual first set. A second set of examples is collected, and the parameters are modified to minimize a combined cost of labeling the first and second sets of examples. The part of the combined cost that represents the cost of the modified parameters applied to the first set is calculated using the approximation function. | 10-22-2009 |
20090263011 | Detection Technique for Digitally Altered Images - Techniques are generally described to determine whether a JPEG image has undergone two compressions. Probabilities can be computed for the first digits of quantized DCT (discrete cosine transform) coefficients from a set of AC (alternate current) modes to detect or determine whether the JPEG image has undergone two compressions. The set of AC modes may include a predetermined number of distinguishable AC modes where a distinguishable AC mode may be an AC mode in which a second quantization step (QS | 10-22-2009 |
20090285473 | METHOD AND APPARATUS FOR OBTAINING AND PROCESSING IMAGE FEATURES - Machine-readable media, methods, apparatus and system for obtaining and processing image features are described. In some embodiments, groups of training features derived from regions of training images may be trained to obtain a plurality of classifiers, each classifier corresponding to each group of training features. The plurality of classifiers may be used to classify groups of validation features derived from regions of validation images to obtain a plurality of weights, wherein each weight corresponds to each region of the validation images and indicates how important the each region of the validation images is. Then, a weight may be discarded from the plurality of weights based upon a certain criterion. | 11-19-2009 |
20090285474 | System and Method for Bayesian Text Classification - A method for classifying text comprises receiving data containing text and parsing a plurality of tokens out of the text. A plurality of metatokens are generated for each token, wherein the metatokens comprise strings of text and groupings of strings of text. The method further comprises calculating a probability that the data falls into a certain category, using the tokens and metatokens. The probability is compared to a threshold value and the data is classified into the certain category if the probability is greater than the threshold value. | 11-19-2009 |
20090290788 | System and method for performing multi-image training for pattern recognition and registration - A system and method for performing multi-image training for pattern recognition and registration is provided. A machine vision system first obtains N training images of the scene. Each of the N images is used as a baseline image and the N−1 images are registered to the baseline. Features that represent a set of corresponding image features are added to the model. The feature to be added to the model may comprise an average of the features from each of the images in which the feature appears. The process continues until every feature that meets a threshold requirement is accounted for. The model that results from the present invention represents those stable features that are found in at least the threshold number of the N training images. The model may then be used to train an alignment/inspection tool with the set of features. | 11-26-2009 |
20090304268 | System and Method for Parallelizing and Accelerating Learning Machine Training and Classification Using a Massively Parallel Accelerator - A method system for training an apparatus to recognize a pattern includes providing the apparatus with a host processor executing steps of a machine learning process; providing the apparatus with an accelerator including at least two processors; inputting training pattern data into the host processor; determining coefficient changes in the machine learning process with the host processor using the training pattern data; transferring the training data to the accelerator; determining kernel dot-products with the at least two processors of the accelerator using the training data; and transferring the dot-products back to the host processor. | 12-10-2009 |
20090310854 | Multi-Label Multi-Instance Learning for Image Classification - Described is a technology by which an image is classified (e.g., grouped and/or labeled), based on multi-label multi-instance data learning-based classification according to semantic labels and regions. An image is processed in an integrated framework into multi-label multi-instance data, including region and image labels. The framework determines local association data based on each region of an image. Other multi-label multi-instance data is based on relationships between region labels of the image, relationships between image labels of the image, and relationships between the region and image labels. These data are combined to classify the image. Training is also described. | 12-17-2009 |
20090310855 | EVENT DETECTION METHOD AND VIDEO SURVEILLANCE SYSTEM USING SAID METHOD - An event detection method for video surveillance systems and a related video surveillance system are described. The method comprises a learning phase, wherein learning images of a supervised area are acquired at different time instants in the absence of any detectable events, and an operating detection phase wherein current images of said area are acquired. The method detects an event by comparing a current image with an image corresponding to a linear combination of a plurality of reference images approximating, or coinciding with, respective learning images. | 12-17-2009 |
20090316982 | TRANSFORMING MEASUREMENT DATA FOR CLASSIFICATION LEARNING | 12-24-2009 |
20090316983 | Real-Time Action Detection and Classification - The present invention relates to a method and system for creating a strong classifier based on motion patterns wherein the strong classifier may be used to determine an action being performed by a body in motion. When creating the strong classifier, action classification is performed by measuring similarities between features within motion patterns. Embodiments of the present invention may utilize candidate part-based action sets and training samples to train one or more weak classifiers that are then used to create a strong classifier. | 12-24-2009 |
20090324060 | LEARNING APPARATUS FOR PATTERN DETECTOR, LEARNING METHOD AND COMPUTER-READABLE STORAGE MEDIUM - A learning apparatus for a pattern detector, which includes a plurality of weak classifiers and detects a specific pattern from input data by classifications of the plurality of weak classifiers, acquires a plurality of data for learning in each of which whether or not the specific pattern is included is given, makes the plurality of weak classifiers learn by making the plurality of weak classifiers detect the specific pattern from the acquired data for learning, selects a plurality of weak classifiers to be composited from the weak classifiers which have learned, and composites the plurality of weak classifiers into one composite weak classifier based on comparison between a performance of the composite weak classifier and performances of the plurality of weak classifiers. | 12-31-2009 |
20100014751 | IMAGE PROCESSING DEVICE, STORAGE MEDIUM, AND IMAGE PROCESSING METHOD - A face candidate of an object having movable ears is detected by a face candidate detection unit from an image of the object, and an ear of the object is detected by an attached site detection unit. The object is then detected from the image by a head portion determination unit in accordance with the detected face candidate and ear. | 01-21-2010 |
20100027875 | AUTOMATED LEARNING FOR PEOPLE COUNTING SYSTEMS - A system, method and program product for providing automated learning for a people counting system. A system is disclosed that includes a grid system for dividing a field of view (FOV) of a captured image data into a set of blocks; an object detection and tracking system for tracking a blob passing through the FOV; and a learning system that maintains person size parameters for each block and updates person size parameters for a selected block when a blob appears in the selected block. | 02-04-2010 |
20100046829 | IMAGE STYLIZATION USING SPARSE REPRESENTATION - A computer-implemented method that includes segmenting a training image into training image patches, where each training image patch is represented by a linear combination of dictionary image patches from an image dictionary, and each dictionary image patch has a sparse representation coefficient. The method includes segmenting a stylized training image into stylized training image patches, where each stylized training image patch is represented by a linear combination of stylized dictionary image patches from a stylized image dictionary, and each stylized dictionary image patch has a sparse representation coefficient. The method also includes training the image dictionary with the training image patches and the stylized image dictionary with the stylized training image patches in a substantially simultaneous manner. The sparse representation coefficient for each training image patch is substantially similar to the sparse representation coefficient for the corresponding stylized training image patch. | 02-25-2010 |
20100080450 | CLASSIFICATION VIA SEMI-RIEMANNIAN SPACES - Described is using semi-Riemannian geometry in supervised learning to learn a discriminant subspace for classification, e.g., labeled samples are used to learn the geometry of a semi-Riemannian submanifold. For a given sample, the K nearest classes of that sample are determined, along with the nearest samples that are in other classes, and the nearest samples in that sample's same class. The distances between these samples are computed, and used in computing a metric matrix. The metric matrix is used to compute a projection matrix that corresponds to the discriminant subspace. In online classification, as a new sample is received, it is projected into a feature space by use of the projection matrix and classified accordingly. | 04-01-2010 |
20100080451 | IMAGE PROCESSING APPARATUS AND COEFFICIENT LEARNING APPARATUS - An image processing apparatus includes a storage unit in which regression coefficient data is stored for each class on the basis of a tap in which a linear feature amount corresponding to a pixel of interest of first image data and a non-linear feature amount determined from the image data are used as elements; a classification unit configured to classify each of linear feature amounts of a plurality of items of input data of the input first image into a predetermined class; a reading unit configured to read the regression coefficient data; and a data generation unit configured to generate data of a second image obtained by making the first image have higher quality by performing a product-sum computation process by using the regression coefficient data read from the reading unit and elements of the tap of each of the plurality of items of input data of the input first image. | 04-01-2010 |
20100080452 | COEFFICIENT LEARNING APPARATUS AND METHOD, IMAGE PROCESSING APPARATUS AND METHOD, PROGRAM, AND RECORDING MEDIUM - A coefficient learning apparatus includes a regression coefficient calculation unit configured to obtain a tap from an image of a first signal; a regression prediction value calculation unit configured to perform a regression prediction computation; a discrimination information assigning unit configured to assign discrimination information to the pixel of interest; a discrimination coefficient calculation unit configured to obtain a tap from the image of the first signal; a discrimination prediction value calculation unit configured to perform a discrimination prediction computation; and a classification unit configured to classify each of the pixels of the image of the first signal into one of the first discrimination class and the second discrimination class. The regression coefficient calculation unit further calculates the regression coefficient using only the pixels classified as the first discrimination class and further calculates the regression coefficient using only the pixel classified as the second discrimination class. | 04-01-2010 |
20100092073 | SYSTEM AND METHOD FOR OBJECT RECOGNITION AND CLASSIFICATION USING A THREE-DIMENSIONAL SYSTEM WITH ADAPTIVE FEATURE DETECTORS - A method including imaging an object in three-dimensions; binning data of the imaged object into three-dimensional regions having a predetermined size; determining a density value p of the data in each bin; inputting the p density values of the bins into a first layer of a computational system including a corresponding processing element for each of the bins; calculating an output O of the processing elements of the computational system while restricting the processing elements to have weights Wc | 04-15-2010 |
20100092074 | FOREGROUND ACTION ESTIMATING APPARATUS AND FOREGROUND ACTION ESTIMATING METHOD - The present invention provides a foreground action estimating apparatus and a foreground action estimating method, wherein the foreground action estimating apparatus includes: a training image inputting means for inputting a foreground image, a background image and an image having the foreground and background images as training images; a basis matrix calculating means for calculating a foreground basis matrix and a background basis matrix by respectively extracting a foreground feature and a background feature from the foreground image and the background image, respectively, and combining the foreground basis matrix and the background basis matrix to obtain a combined basis matrix; a feature suppressing means for calculating the feature coefficients of the training images in accordance with the combined basis matrix obtained by the basis matrix calculating means so as to obtain image features of the background-feature-suppressed training images; and a foreground action information acquiring means for estimating foreground action information in accordance with a feature mapping matrix from the image feature to an action information set, by using the background-feature-suppressed image features. | 04-15-2010 |
20100092075 | Method of directed pattern enhancement for flexible recognition - A directed pattern enhancement method receives a learning image and pattern enhancement directive. Pattern enhancement learning is performed using the learning image and the pattern enhancement directive to generate pattern enhancement recipe. An application image is received and a pattern enhancement application is performed using the application image and the pattern enhancement recipe to generate pattern enhanced image. A recognition thresholding is performed using the pattern enhanced image to generate recognition result. The pattern enhancement directive consists of background directive, patterns to enhance directive, and patterns to suppress directive. An update learning method performs pattern enhancement progressive update learning. | 04-15-2010 |
20100128975 | METHOD AND SYSTEM FOR OBJECT RECOGNITION BASED ON A TRAINABLE DYNAMIC SYSTEM - A system for object recognition in which a multi-dimensional scanner generates a temporal sequence of multi-dimensional output data of a scanned object. That data is then coupled as an input signal to a trainable dynamic system. The system exemplified by a general-purpose recurrent neural network is previously trained to generate an output signal representative of the class of the object in response to a temporal sequence of multi-dimensional data. | 05-27-2010 |
20100150432 | METHOD AND MACHINE FOR DIGITALLY CATALOGUING ARTICLES - Method for digitally cataloguing articles, in which a revolving support supports an article and turns it around a vertical axis of rotation, a plurality of images of the article are acquired synchronously with its rotation y two digital cameras positioned at respective observation points on a fixed structure, the acquired images of the article are filed in an image database, the filed images are made-up together with information pertinent to the article extracted from a stock control program of the articles to catalogue, and the thus made-up pages are published on the Internet. | 06-17-2010 |
20100158356 | SYSTEM AND METHOD FOR IMPROVED CLASSIFICATION - A system and method for improved classification. A first classifier is trained using a first process running on at least one computing device using a first set of training images relating to a class of images. A set of additional images are selected using the first classifier from a source of additional images accessible to the computing device. The first set of training images and the set of additional images are merged using the computing device to create a second set of training images. A second classifier is trained using a second process running on the computing device using the second set of training images. A set of unclassified images are classified using the second classifier thereby creating a set of classified images. The first classifier and the second classifier employ different classification methods. | 06-24-2010 |
20100172573 | Distinguishing Colors of Illuminated Objects Using Machine Vision - System and method for distinguishing colors of illuminated objects using machine vision. A color-balanced image that includes at least one lit area is received, as well as an indication of a region of interest that includes one of the one or more lit areas. A mask image is generated based on the region of interest. A color-balanced image of the region of interest is generated by masking the color-balanced image with the mask image, and a plurality of image attributes for the region of interest is determined by analyzing the color-balanced image of the region of interest. A color is determined based on the plurality of image attributes using a trained classifier, and the determined color stored, e.g., in a memory medium. | 07-08-2010 |
20100177956 | SYSTEMS AND METHODS FOR SCALABLE MEDIA CATEGORIZATION - Systems and methods for automating digital file classification are described. The systems and methods include generating a plurality of classifiers from a plurality of first features of a plurality of first digital files, each of the plurality of first digital files having one or more associated annotations. A plurality of second features extracted from a plurality of second digital files is sorted according to the plurality of classifiers. A distance vector is determined between the second features and respective first features for the corresponding ones of the classifiers and the determined distances are ranked. A subset of matched files is selected based on the ranking. The subset of matched files correspond to respective one or more associated annotations. One or more annotations associated with the subset of matched files are associated to subsequently received digital files using the corresponding ones of the classifiers. | 07-15-2010 |
20100183218 | OBJECT DETERMINING DEVICE AND PROGRAM THEREOF - An object determining device includes imaging means for obtaining an image of the object, likelihood value calculating means for calculating a first likelihood value for the object shown in the image by use of the image obtained by the imaging means and a machine learning system and for calculating a second likelihood value for the object shown in the image by use of the image obtained by the imaging means and another machine learning system, the first likelihood value indicating a level of likelihood that the object is wearing the covering and the second likelihood value indicating a level of likelihood that the object is not wearing the covering, and determining means for determining whether or not the object, shown in the image obtained by the imaging means, is wearing a covering, on the basis of a ratio between the first likelihood value and the second likelihood value. | 07-22-2010 |
20100202681 | DETECTING DEVICE OF SPECIAL SHOT OBJECT AND LEARNING DEVICE AND METHOD THEREOF - The invention discloses a detecting device for specific subjects and a learning device and method thereof. The detecting device for specific subjects includes an input unit, one or more strong classifying units, a storage unit and a judging unit, wherein the input unit is used for inputting images to be detected; the strong classifying units are used for carrying out strong classification to the image, each strong classifying unit includes one or more weak classifying units, and the weak classifying unit carries out weak classification to the image with a weak classifying template; the storage unit stores the weak classifying template used by the weak classifying unit; and the judging unit judges whether or not the image contains specific subjects according to the classification result of the strong classifying unit. The detecting device for specific subjects also includes an incremental sample input unit and a learning unit, wherein the incremental sample input unit is used for inputting data for incremental learning, namely for inputting an incremental learning sample, which is data undetected and wrongly detected by the detecting device or other detecting devices for specific subjects; the learning unit is used for updating the weak classifying template stored in the storage unit according to the incremental learning sample inputted by the incremental sample input unit. | 08-12-2010 |
20100208983 | LEARNING DEVICE, LEARNING METHOD, IDENTIFICATION DEVICE, IDENTIFICATION METHOD, AND PROGRAM - A learning device includes a feature-point extracting section extracting feature points from a generation image, a feature-point feature-quantity extracting section extracting feature-point feature-quantities representing features of the feature points, a total-feature-quantity generating section generating a total feature quantity represented by a multi-dimensional vector, and an identifier generating section generating an identifier using the total feature quantity and a true label indicating whether or not the generation image is a positive image or a negative image. | 08-19-2010 |
20100215254 | Self-Learning Object Detection and Classification Systems and Methods - A method of object classification based upon fusion of a remote sensing system and a natural imaging system is provided. The method includes detecting an object using the remote sensing system. An angle of view of a video camera of the natural imaging system is varied. An image including the object is generated using the natural imaging system. The natural imaging system may zoom in on the object. The image represented in either pixel or transformed space is compared to a plurality of templates via a competition based neural network learning algorithm. Each template has an associated label determined statistically. The template with a closest match to the image is determined. The image may be assigned the label associated with the relative location of the object, the relative speed of the object, and the label of the template determined statistically to be the closest match to the image. | 08-26-2010 |
20100215255 | Iterative Data Reweighting for Balanced Model Learning - Aspects of the present invention include systems and methods for forming generative models, for utilizing those models, or both. In embodiments, an object model fitting system can be developed comprising a 3D active appearance model (AAM) model. The 3D AAM comprises an appearance model comprising a set of subcomponent appearance models that is constrained by a 3D shape model. In embodiments, the 3D AAM may be generated using a balanced set of training images. The object model fitting system may further comprise one or more manifold constraints, one or more weighting factors, or both. Applications of the present invention include, but are not limited to, modeling and/or fitting face images, although the teachings of the present invention can be applied to modeling/fitting other objects. | 08-26-2010 |
20100215256 | METHOD AND DEVICE FOR MAINTAINING IMAGE BACKGROUND BY MULTIPLE GAUSSIAN MODELS - A method maintaining an image background by multiple Gaussian models utilized to a device includes the following steps. First, the device captures an image frame having pixels to obtain background information, and then calculates the background information to establish a primary Gaussian model. Next, the device captures continuous image frames in a time period to obtain and calculate graphic information for establishing a secondary Gaussian model, and then repeates the steps to establish multiple secondary Gaussian models. Finally, the device compares two secondary Gaussian models, and then updates learning for the primary Gaussian model by the secondary Gaussian model if the graphic information of the secondary Gaussian models are attributable to the background information, or maintains the background information of the primary Gaussian model without updating the learning if anyone of the graphic information of the two secondary Gaussian models is unattributable to the background information. | 08-26-2010 |
20100215257 | CAPTURING AND RECOGNIZING HAND POSTURES USING INNER DISTANCE SHAPE CONTEXTS - A system, method, and computer program product for recognizing hand postures are described. According to one aspect, a set of training images is provided with labels identifying hand states captured in the training images. Inner Distance Shape Context (IDSC) descriptors are determined for the hand regions in the training images, and fed into a Support Vector Machine (SVM) classifier to train it to classify hand shapes into posture classes. An IDSC descriptor is determined for a hand region in a testing image, and classified by the SVM classifier into one of the posture classes the SVM classifier was trained for. The hand posture captured in the testing image is recognized based on the classification. | 08-26-2010 |
20100220922 | LEARNING APPARATUS AND OBJECT DETECTING APPARATUS - Feature values calculated from a peripheral image area of feature points extracted in a detection target object in a training image each are labeled with a label indicating a class of the detection target object, feature values calculated from a peripheral image area of feature points of a non detection target object in the training image each are labeled with a label indicating the non detection target object, voting positions in a parameter space are calculated by relative positions of the feature points of the detection target object from the detection target object on the training image, and a first classifier is learned using the labeled feature values extracted in the training image so that a class distribution is concentrated and the voting positions in the parameter space are concentrated. | 09-02-2010 |
20100226564 | FRAMEWORK FOR IMAGE THUMBNAILING BASED ON VISUAL SIMILARITY - An apparatus and method for detecting a region of interest in an image are disclosed. Image representations for a set of images that have been manually annotated with regions of interest are stored, along with positive and negative representations of each image which are similarly derived to the image representations except that they are based on features extracted from patches within the region of interest and outside it, respectively. For an original image for which a region of interest is desired, the stored information for K similar images is automatically retrieved and used to train a classifier. The trained classifier provides, for each patch of the original image, a probability of being in a region of interest, based extracted features of the patch (represented, for example, as a Fisher vector), which can be used to determine a region of interest in the original image. | 09-09-2010 |
20100232685 | IMAGE PROCESSING APPARATUS AND METHOD, LEARNING APPARATUS AND METHOD, AND PROGRAM - An image processing apparatus includes: an edge intensity detecting unit configured to detect the edge intensity of an image in increments of blocks having a predetermined size; a parameter setting unit configured to set an edge reference value used for extraction of an edge point that is a pixel used for detection of the blurred degree of the image based on a dynamic range that is difference between the maximum value and the minimum value of the edge intensities; and an edge point extracting unit configured to extract a pixel as the edge point with the edge intensity being equal to or greater than the edge reference value, and also the pixel value of a pixel within a block being included in an edge block that is a block within a predetermined range. | 09-16-2010 |
20100232686 | HIERARCHICAL DEFORMABLE MODEL FOR IMAGE SEGMENTATION - Described herein is a technology for facilitating deformable model-based segmentation of image data. In one implementation, the technology includes receiving training image data ( | 09-16-2010 |
20100246940 | METHOD OF GENERATING HDR IMAGE AND ELECTRONIC DEVICE USING THE SAME - A method of generating a high dynamic range image and an electronic device using the same are described. The method includes loading a brightness adjustment model created by a neural network algorithm; obtaining an original image; acquiring a pixel characteristic value, a first characteristic value in a first direction, and a second characteristic value in a second direction of the original image; and generating an HDR image through the brightness adjustment model according to the pixel characteristic value, the first characteristic value, and the second characteristic value of the original image. The electronic device includes a brightness adjustment model, a characteristic value acquisition unit, and a brightness adjustment procedure. The electronic device acquires a pixel characteristic value, a first characteristic value, and a second characteristic value of an original image through the characteristic value acquisition unit, and generates an HDR image from the original image through the brightness adjustment model. | 09-30-2010 |
20100254595 | GRAPHIC RECOGNITION DEVICE, GRAPHIC RECOGNITION METHOD, AND GRAPHIC RECOGNITION PROGRAM - A graphic recognition device, method, and recognition program recognize graphics without being influenced by an image shadow area. Image input unit acquires the image of the outside environment of a vehicle using a vehicle mounted camera. A light source location information acquiring unit calculates location of a light source such as the sun using the acquired image. User vehicle shape acquiring unit and other vehicle shape acquiring unit generate shape information for the vehicles indicating the location of points forming vehicle contours. Shadow area calculating unit calculates, on the basis of both vehicles' shape information, the object shape information and the light source location information, the location coordinates of the shadow area, and converts the location coordinates into two-dimensional coordinates to the shadow/non-shadow area emphasis flag recognizing unit, which recognizes the flag in the image by judging the presence/absence of the recognition object in each shadow and non-shadow area specified. | 10-07-2010 |
20100254596 | METHOD AND SYSTEM FOR GENERATING AN ENTIRELY WELL-FOCUSED IMAGE OF A LARGE THREE-DIMENSIONAL SCENE - A method and system for generating an entirely well-focused image of a three-dimensional scene. The method comprises the steps of a) learning a prediction model including at least a focal depth probability density function (PDF), h(k), for all depth values k, from historical tiles of the scene; b) predicting the possible focal surfaces in subsequent tiles of the scene by applying the prediction model; c) for each value of k, examining h(k) such that if h(k) is below a first threshold, no image is acquired at the depth k′ for said one tile; and if h(k) is above or equal to a first threshold, one or more images are acquired in a depth range around said value of k for said one tile; and d) processing the acquired images to generate a pixel focus map for said one tile. | 10-07-2010 |
20100272349 | REAL-TIME ANNOTATION OF IMAGES IN A HUMAN ASSISTIVE ENVIRONMENT - A method, information processing system, and computer program storage product annotate video images associated with an environmental situation based on detected actions of a human interacting with the environmental situation. A set of real-time video images are received that are captured by at least one video camera associated with an environment presenting one or more environmental situations to a human. One or more user actions made by the human that is associated with the set of real-time video images with respect to the environmental situation are monitored. A determination is made, based on the monitoring, that the human driver has one of performed and failed to perform at least one action associated with one or more images of the set of real-time video images. The one or more images of the set of real-time video images are annotated with a set of annotations. | 10-28-2010 |
20100272350 | METHODS AND APPARATUS TO PERFORM IMAGE CLASSIFICATION BASED ON PSEUDORANDOM FEATURES - Example methods and apparatus to perform image classification based on pseudorandom features are disclosed. A disclosed example method includes generating first and second pseudorandom numbers, extracting a first feature of an image based on the first and second pseudorandom numbers, and determining a classification for the image based on the first extracted feature. | 10-28-2010 |
20100272351 | INFORMATION PROCESSING APPARATUS AND METHOD FOR DETECTING OBJECT IN IMAGE DATA - Learning is sequentially executed with respect to weak discriminators based on learning data held in a storage device. Upon learning, an evaluation value for the weak discriminator is calculated. It is discriminated, based on a shift of the evaluation value, whether or not the learning is overlearning. If it is discriminated that the learning is overlearning, new learning data is added. Thus, the overlearning is easily detected and the learning is efficiently executed. | 10-28-2010 |
20100290700 | INFORMATION PROCESSING DEVICE AND METHOD, LEARNING DEVICE AND METHOD, PROGRAMS, AND INFORMATION PROCESSING SYSTEM - An information processing device including an extraction unit and a detection unit. If both a parameter set extracting features from an image and a classifier performing predetermined classification by using the extracted features are statistically learned in advance, the extraction unit extracts features of a recognition target object from an input image by using the parameter set, and the detection unit performs the predetermined classification by using the classifier, which uses the features extracted by the extraction unit, and, on the basis of the result of the classification, determines whether or not the object is included in the input image. | 11-18-2010 |
20100296728 | Discrimination Apparatus, Method of Discrimination, and Computer Program - A discrimination apparatus includes: a feature-quantity extraction section extracting a feature quantity from an object of discrimination; and a discriminator including a plurality of weak discriminators expressed as a Bayesian network having each node to which a corresponding one of two or more of the feature quantities input from the feature-quantity extraction section is allocated and a combiner combining individual discrimination results of the object of discrimination by the plurality of weak discriminators. | 11-25-2010 |
20100303343 | METHOD FOR FACE RECOGNITION AND SYNTHESIS - A method of recognizing an object in an image is provided, the method comprises the following steps. The image having the object is provided, and principal traits of the object are encoded in order to generate a first trait code. The first trait code is compared with data stored in a database so as to obtain a plurality of differences. A minimum of the plurality of differences is found. This method can be applied to synthesize human faces. | 12-02-2010 |
20100310156 | Image Data Processing And Arrangements Therefor - Image data processing is facilitated. According to an example embodiment, image data is processed using photometric similarity and, where appropriate, a classification of sample pixels for the image data. In some applications, a trained bilateral filter function is used with a filter coefficient selected for a particular classification of image data to filter artifacts in the image data. | 12-09-2010 |
20100310157 | Apparatus and method for video sensor-based human activity and facial expression modeling and recognition - An apparatus and method for human activity and facial expression modeling and recognition are based on feature extraction techniques from time sequential images. The human activity modeling includes determining principal components of depth and/or binary shape images of human activities extracted from video clips. Independent Component Analysis (ICA) representations are determined based on the principal components. Features are determined through Linear Discriminant Analysis (LDA) based on the ICA representations. A codebook is determined using vector quantization. Observation symbol sequences in the video clips are determined. And human activities are learned using the Hidden Markov Model (HMM) based on status transition and an observation matrix. | 12-09-2010 |
20100310158 | Method And Apparatus For Training Classifier, Method And Apparatus For Image Recognition - Embodiments of the present invention provide a method and apparatus for training an image classifier. The method includes: A. dividing a set of training images for classifier training into a positive-example sample set and at least two negative-example sample sets; B. determining, for each negative-example sample set, a feature set for differentiating the positive-example sample set from the negative-example sample set; and C. performing training using each feature set determined to obtain a classifier. This invention also provides a method and apparatus for image recognition utilizing the image classifier. | 12-09-2010 |
20100329544 | INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, AND PROGRAM - An information processing apparatus includes the following elements. A learning unit is configured to perform Adaptive Boosting Error Correcting Output Coding learning using image feature values of a plurality of sample images each being assigned a class label to generate a multi-class classifier configured to output a multi-dimensional score vector corresponding to an input image. A registration unit is configured to input a register image to the multi-class classifier, and to register a multi-dimensional score vector corresponding to the input register image in association with identification information about the register image. A determination unit is configured to input an identification image to be identified to the multi-class classifier, and to determine a similarity between a multi-dimensional score vector corresponding to the input identification image and the registered multi-dimensional score vector corresponding to the register image. | 12-30-2010 |
20110019907 | METHOD FOR IDENTIFYING MARKED IMAGES USING STATISTICAL MOMENTS BASED AT LEAST IN PART ON A JPEG ARRAY - Briefly, embodiments of a method of identifying marked images, in which higher order statistical moments based at least in part on a JPEG array are employed, is described | 01-27-2011 |
20110026811 | IMAGE PROCESSING APPARATUS AND METHOD, DATA PROCESSING APPARATUS AND METHOD, AND PROGRAM AND RECORDING MEDIUM - The image processing apparatus and method, and the program and the recording medium according to the present invention can make the coefficient vector into high precision by noise elimination or correction utilizing the mutual correlation of the divided image areas in the intermediate eigenspace, and allows relaxation of the input condition and robustness. The high correlation in the divided image areas in the intermediate eigenspace can reduce the divided image areas to be processed, and actualize reduction in processing load and enhancement of the processing speed. | 02-03-2011 |
20110026812 | OBJECT DETECTION APPARATUS - In order to improve object detection precision, an object detection apparatus includes a posterior probability calculation portion utilizing an occurrence probability of a background and a foreground acquired by utilizing a characteristic quantity extracted from each pixel of an input image and a probability density function, a posterior probability of the previous background and foreground, and a conditional probability indicating a relation of an event (background or foreground) to the vicinity of an attentive pixel in a spatial direction and a relation of an event to the vicinity of the attentive pixel in a temporal direction so as to calculate a posterior probability of the background and the foreground from a probability model utilizing a tendency that the same event appears together in the vicinity of the attentive pixel in the spatial and temporal directions; and an object determination portion for determining an object from comparison between the posterior probabilities of the background and the foreground. | 02-03-2011 |
20110044534 | HIERARCHICAL CLASSIFIER FOR DATA CLASSIFICATION - Described herein is a framework for constructing a hierarchical classifier for facilitating classification of digitized data. In one implementation, a divergence measure of a node of the hierarchical classifier is determined. Data at the node is divided into at least two child nodes based on a splitting criterion to form at least a portion of the hierarchical classifier. The splitting criterion is selected based on the divergence measure. If the divergence measure is less than a predetermined threshold value, the splitting criterion comprises a divergence-based splitting criterion which maximizes subsequent divergence after a split. Otherwise, the splitting criterion comprises an information-based splitting criterion which seeks to minimize subsequent misclassification error after the split. | 02-24-2011 |
20110052046 | SYSTEM AND METHOD FOR VISUAL SEARCHING OF OBJECTS USING LINES - Disclosed is method of visual search for objects that include straight lines. A two-step process is used, which includes detecting straight line segments in an image. The lines are generally characterized by their length, midpoint location, and orientation. Hypotheses that a particular straight line segment belongs to a known object are generated and tested. The set of hypotheses is constrained by spatial relationships in the known objects. The speed and robustness of the method and apparatus disclosed makes it immediately applicable to many computer vision applications. | 03-03-2011 |
20110058734 | CLASSIFICATION OF IMAGES AS ADVERTISEMENT IMAGES OR NON-ADVERTISEMENT IMAGES - An advertisement image classification system trains a binary classifier to classify images as advertisement images or non-advertisement images and then uses the binary classifier to classify images of web pages as advertisement images or non-advertisement images. During a training phase, the classification system generates training data of feature vectors representing the images and labels indicating whether an image is an advertisement image or a non-advertisement Image. The classification system trains a binary classifier to classify Images using training data. During a classification phase, the classification system inputs a web page with an image and generates a feature vector for the image. The classification system then applies the trained binary classifier to the feature vector to generate a score indicating whether the image is an advertisement image or a non-advertisement image. | 03-10-2011 |
20110064301 | TEXTUAL ATTRIBUTE-BASED IMAGE CATEGORIZATION AND SEARCH - Techniques and systems for providing textual attribute-based image categorization and search are disclosed herein. In some aspects, images may be analyzed to identify a category of an image, or portion thereof. Additional textual attributes may be identified and associated with the image. In various aspects, the categories may be types of sky sceneries. Categorized images may be searched based on the categories and/or attributes. In further aspects, a user interface may provide an intuitive arrangement of the images for user navigation and selection. The user interface may also provide a simplified presentation and search of the categorized images. Images selected from user interface may be used to replace or modify features of an existing target image. | 03-17-2011 |
20110064302 | RECOGNITION VIA HIGH-DIMENSIONAL DATA CLASSIFICATION - A method is disclosed for recognition of high-dimensional data in the presence of occlusion, including: receiving a target data that includes an occlusion and is of an unknown class, wherein the target data includes a known object; sampling a plurality of training data files comprising a plurality of distinct classes of the same object as that of the target data; and identifying the class of the target data through linear superposition of the sampled training data files using l | 03-17-2011 |
20110064303 | Object Recognition Using Textons and Shape Filters - Given an image of structured and/or unstructured objects, semantically meaningful areas are automatically partitioned from the image, each area labeled with a specific object class. Shape filters are used to enable capturing of some or all of the shape, texture, and/or appearance context information. A shape filter comprises one or more regions of arbitrary shape, size, and/or position within a bounding area of an image, paired with a specified texton. A texton comprises information describing the texture of a patch of surface of an object. In a training process a sub-set of possible shape filters is selected and incorporated into a conditional random field model of object classes. The conditional random field model is then used for object detection and recognition. | 03-17-2011 |
20110075917 | ESTIMATING AESTHETIC QUALITY OF DIGITAL IMAGES - A method for estimating the aesthetic quality of an input digital image comprising using a digital image processor for performing the following: determining one or more vanishing point(s) associated with the input digital image by automatically analyzing the digital image; computing a compositional model from at least the positions of the vanishing point(s); and producing an aesthetic quality parameter for the input digital image responsive to the compositional model, wherein the aesthetic quality parameter is an estimate for the aesthetic quality of the input digital image. | 03-31-2011 |
20110075918 | Method and system for learning object recognition in images - In a first exemplary embodiment of the present invention, an automated, computerized method for learning object recognition in an image is provided. According to a feature of the present invention, the method comprises the steps of providing a training set of standard images, calculating intrinsic images corresponding to the standard images and building a classifier as a function of the intrinsic images. | 03-31-2011 |
20110075919 | Techniques for Enabling or Establishing the Use of Face Recognition Algorithms - Embodiments described herein facilitate or enhance the implementation of image recognition processes which can perform recognition on images to identify objects and/or faces by class or by people. | 03-31-2011 |
20110081073 | Methods And Logic For Autonomous Generation Of Ensemble Classifiers, And Systems Incorporating Ensemble Classifiers - In one embodiment, a method for generating an ensemble classifier may include transforming multidimensional training data into a plurality of response planes. Each of the response planes includes a set of confidence scores. The response planes are transformed into a plurality of binary response planes. Each of the binary response planes include a set of binary scores corresponding to one of the confidence scores. Combinations of the binary response planes are transformed into sets of diversity metrics according to a diversity measure. A metric is selected from the sets of diversity metrics. A predicted performance of a child combination of the recognition algorithms corresponding to the combinations is generated. The predicted performance is based at least in part upon the metrics. Parent recognition algorithms are selected from the recognition algorithms based at least in part upon the predicted performance. The ensemble classifier is generated and includes the parent recognition algorithms. | 04-07-2011 |
20110091097 | APPARATUS OF LEARNING RECOGNITION DICTIONARY, AND METHOD OF LEARNING RECOGNITION DICTIONARY - There are provided a characteristic obtaining unit configured to obtain a subject characteristic including a characteristic of a subject, an image processing unit configured to generate a duplicate subject image by performing an image process to an image of the subject according to the subject characteristic obtained by the characteristic obtaining unit, and a learning unit configured to learn a matching dictionary by using the duplicate subject image generated by the image processing unit. Thus, it is possible to reduce the number of subject images necessary for the learning. | 04-21-2011 |
20110091098 | System and Method for Detecting Text in Real-World Color Images - A method and apparatus for detecting text in real-world images comprises calculating a cascade of classifiers, the cascade comprising a plurality of stages, each stage including one or more weak classifiers, the plurality of stages organized to start out with classifiers that are most useful for ruling out non-text regions, and removing regions classified as non-text regions from the cascade prior to completion of the cascade, to further speed up processing. | 04-21-2011 |
20110103682 | MULTI-MODALITY CLASSIFICATION FOR ONE-CLASS CLASSIFICATION IN SOCIAL NETWORKS - A classification apparatus, method, and computer program product for multi-modality classification are disclosed. For each of a plurality of modalities, the method includes extracting features from objects in a set of objects. The objects include electronic mail messages. A representation of each object for that modality is generated, based on its extracted features. At least one of the plurality of modalities is a social network modality in which social network features are extracted from a social network implicit in the set of electronic mail messages. A classifier system is trained based on class labels of a subset of the set of objects and on the representations generated for each of the modalities. With the trained classifier system, labels are predicted for unlabeled objects in the set of objects. | 05-05-2011 |
20110123101 | INDOOR-OUTDOOR DETECTOR FOR DIGITAL CAMERAS - An indoor-outdoor detection method includes constructing a first indoor-outdoor detector; and constructing a second indoor-outdoor detector. A normalized brightness of a subject image is determined and a comparison result is generated based on the determined normalized brightness and a threshold brightness value. One of the first detector or the second detector is selectively applied to the subject image based on the comparison result, and a detection result is generated. Image signal processing is performed on the subject image based on the detection result. | 05-26-2011 |
20110123102 | IMAGE PROCESSING DEVICE, METHOD THEREOF, AND STORAGE MEDIUM STORING IMAGE PROCESSING PROGRAM - An image processing device includes a dictionary data storage unit to store dictionary data regarding features that a plurality of objects has, an arithmetic unit to compute feature data of an input image based on information of the input image that includes an object with a specific feature among the plurality of objects, and a calculation unit to calculate a parameter for adjusting the dictionary data regarding the object with the specific feature based on the feature data and the dictionary data. | 05-26-2011 |
20110129145 | DETECTING FACIAL SIMILARITY BASED ON HUMAN PERCEPTION OF FACIAL SIMILARITY - Similar faces may be determined within images based on human perception of facial similarity. The user may provide an image including a query face to which the user wishes to find faces that are similar. Similar faces may be determined based on similarity information. Similarity information may be generated from information related to a human perception of facial similarity. Images that include faces determined to be similar, based on the similarity information, may be provided to the user as search result images. The user then may provide feedback to indicate the user's perception of similarity between the query face and the search result images. | 06-02-2011 |
20110135191 | APPARATUS AND METHOD FOR RECOGNIZING IMAGE BASED ON POSITION INFORMATION - According to the present invention, the amount of computation required for image recognition processing can be reduced by extracting only image recognition learning information for an object that may appear in a region having the geographical property of a current position and comparing the image recognition learning information with ambient-image information. | 06-09-2011 |
20110142330 | IMAGE PROCESSING APPARATUS AND METHOD - An image processing apparatus and an image processing method, the image processing apparatus including: an image input unit which receives an image; and an image processing unit which generates reference data on the basis of a plurality of learning images classified into a plurality of first classes according to a noise characteristic and a plurality of second classes according to an image characteristic, and which performs scaling for the received image on the basis of the generated reference data. | 06-16-2011 |
20110150323 | CATEGORIZATION QUALITY THROUGH THE COMBINATION OF MULTIPLE CATEGORIZERS - A system categorizes one or more objects based at least in part upon one or more characteristics associated therewith. A first classifier includes a rule set to determine if each of the one or more objects meets or exceeds a quality threshold. A second classifier, orthogonal to the first classifier, includes a rule set to determine if each of the one or more objects meets or exceeds a quality threshold. In one embodiment, the quality threshold associated with the first classifier and the quality threshold associated with the second classifier are less than a predetermined target threshold. The result for each object of the first classifier is compared to the result of the second classifier. The object is categorized if the result of the first classifier and the result of the second classifier match. The object is uncategorized if the result of the first classifier does not match the result of the second classifier. | 06-23-2011 |
20110150324 | METHOD AND APPARATUS FOR RECOGNIZING AND LOCALIZING LANDMARKS FROM AN IMAGE ONTO A MAP - Method and apparatus for recognizing landmark buildings in an image and then locating the recognized landmark buildings onto a map together with related information wherein a first database is employed to store models formed by mathematical set descriptions of landmark buildings which are learned from a set of training images of a model-learning module captured by an imaging device for each building, and a second database is employed to store the related information of each landmark building. The model of each landmark building is represented as a set of features and the geometric relationship between them by clustering the salient features extracted from a set of training images of the landmark building. | 06-23-2011 |
20110150325 | Visual Object Appearance Modelling Using Image Processing - A computer-implemented method of generating a model from a set of images. The method comprises processing a plurality of data items, each data item representing an image of said set of images, to determine variability between said plurality of data items; and generating model data representing said model based upon said data items and said variability, wherein the influence of each of said data items upon the generated model is determined by a relationship between a respective one of said data items and said variability. | 06-23-2011 |
20110158510 | BIOLOGICALLY-INSPIRED METADATA EXTRACTION (BIME) OF VISUAL DATA USING A MULTI-LEVEL UNIVERSAL SCENE DESCRIPTOR (USD) - A computer vision system provides a universal scene descriptor (USD) framework and methodology for using the USD framework to extract multi-level semantic metadata from scenes. The computer vision system adopts the human vision system principles of saliency, hierarchical feature extraction and hierarchical classification to systematically extract scene information at multiple semantic levels. | 06-30-2011 |
20110158511 | METHOD AND APPARATUS FOR FILTERING RED AND/OR GOLDEN EYE ARTIFACTS - Processing method of a digital image to filter red and/or golden eye artifacts, the digital image comprising a plurality of pixel each comprising at least one digital value represented on a plurality of bits, the method comprising: a step of selecting at least one patch of pixels of the digital image comprising pixels potentially representative of a red and/or golden eye artifact; a step of classifying the at least one patch of pixels as “eye” or “non-eye”; a step of filtering said potentially representative pixels if said patch of pixels is classified as “eye”; wherein the classifying step comprises the operations of: converting the digital values of said patch of pixels into a Gray Code representation, overall obtaining a plurality of bit maps from said patch of pixels, each bit map being associated with a respective bit of said Gray Code; an operation of individually comparing said bit maps with corresponding bit map models belonging to a patch classifier produced by a statistical analysis of bit maps obtained by converting patches of pixels of digital images containing or not red and/or golden eye artifacts into said Gray Code representation. | 06-30-2011 |
20110164812 | METHOD, APPARATUS AND SYSTEM FOR ORIENTING A DISORIENTED IMAGE - A method, apparatus and system for orienting a disoriented image, and a method, apparatus and system for training a plurality of Gaussian mixture models (GMMs) to orient the disoriented image are provided. The method of training the plurality of GMMs includes: obtaining a plurality of color and texture features from the disoriented image; selecting a plurality of discriminative features from the color and texture features; calculating probabilities of each of the GMMs orienting the disoriented image, where each of the GMMs represents one of a plurality of rotation classes, and each of the rotation classes represents a rotation angle that is a multiple of a right angle. Furthermore, the system includes an electronic device that includes an embedded platform including a processor which processes the disoriented image. | 07-07-2011 |
20110170768 | Image segregation system with method for handling textures - In a first exemplary embodiment of the present invention, an automated, computerized method is provided for processing an image. According to a feature of the present invention, the method comprises the steps of providing an image file depicting an image, in a computer memory, assembling a feature vector for the image file, the feature vector containing information regarding a likelihood that a selected pair of regions of the image file are of a same intrinsic characteristic, for example, a same texture, providing a classifier derived from a computer learning technique, computing a classification score for the selected pair of regions of the image file, as a function of the feature vector and the classifier and classifying the regions as being of the same intrinsic characteristic, as a function of the classification score. | 07-14-2011 |
20110170769 | CLASSIFIER LEARNING IMAGE PRODUCTION PROGRAM, METHOD, AND SYSTEM - A classifier learning image production program, method, and system are provided which are capable of efficiently acquiring learning images to be employed in development of a discrimination application, or more particularly, efficiently acquiring initial learning images to be employed in an early stage of development of a discrimination algorithm. A classifier learning image production program allows a computer to execute the steps of inputting an image; detecting a discrimination area from the inputted image, acquiring plural detected data, and recording the detected data in a storage device; integrating the plural detected data to obtain learning image candidate information, and recording the learning image candidate information as the detected data in the storage device; clipping plural learning images from the inputted images, and recording the plural learning images as learning image data in the storage device; classifying the learning images into one or more sets; and displaying the learning images on a display device. | 07-14-2011 |
20110176724 | Content-Aware Ranking for Visual Search - This document describes techniques that utilize a learning method to generate a ranking model for use in image search systems. The techniques leverage textual information and visual information simultaneously when generating the ranking model. The tools are further configured to apply the ranking model responsive to receiving an image search query. | 07-21-2011 |
20110176725 | LEARNING APPARATUS, LEARNING METHOD AND PROGRAM - A learning apparatus includes a learning section which learns, according as a learning image used for learning a discriminator for discriminating whether a predetermined discrimination target is present in an image is designated from a plurality of sample images by a user, the discriminator using a random feature amount including a dimension feature amount randomly selected from a plurality of dimension feature amounts included in an image feature amount indicating features of the learning image. | 07-21-2011 |
20110182500 | CONTEXTUALIZATION OF MACHINE INDETERMINABLE INFORMATION BASED ON MACHINE DETERMINABLE INFORMATION - A system for contextualizing machine indeterminable information based on machine determinable information may include a memory, an interface, and a processor. The memory may store an electronic document image which may include information determinable by a machine and information indeterminable by a machine The processor may be operative to receive, via the interface, the electronic document image. The processor may determine the machine determinable information of the electronic document image and may identify the machine indeterminable information of the electronic document image. The processor may contextualize the machine indeterminable information based on the machine determinable information. The processor may present the contextualized machine indeterminable information to the user to facilitate interpretation thereof. In response thereto, the processor may receive, via the interface, data representative of a user determination associated with the machine indeterminable information. | 07-28-2011 |
20110182501 | METHOD FOR RECOGNIZING SHAPES AND SYSTEM IMPLEMENTING SAID METHOD - The invention includes a method for recognizing shapes using a preprocessing mechanism that decomposes a source signal into basic components called atoms and a recognition mechanism that is based on the result of the decomposition performed by the preprocessing mechanism. In the method, the preprocessing mechanism includes at least one learning phase culminating in a set of signals called kernels, the kernels being adapted to minimize a cost function representing the capacity of the kernels to correctly reconstruct the signals from the database while guaranteeing a sparse decomposition of the source signal while using a database of signals representative of the source to be processed and a coding phase for decomposing the source signal into atoms, the atoms being generated by shifting of the kernels according to their index, each of the atoms being associated with a decomposition coefficient. The invention also includes a shape recognition system for implementing the method. | 07-28-2011 |
20110188742 | RECOMMENDING USER IMAGE TO SOCIAL NETWORK GROUPS - A method of recommending social group(s) for sharing one or more user images, includes using a processor for acquiring the one or more user images and their associated metadata; acquiring one or more group images from the social group(s) and their associated metadata; computing visual features for the user images and the group images; and recommending social group(s) for the one of more user images using both the visual features and the metadata. | 08-04-2011 |
20110188743 | IMAGE PROCESSING APPARATUS, IMAGE PROCESSING METHOD, IMAGE PROCESSING SYSTEM, AND RECORDING MEDIUM - An image processing apparatus applying image processing corresponding to a predetermined imaging region of a subject of an image obtained by capturing the imaging region to the image includes an acquisition unit configured to acquire classification information obtained by classifying a plurality of imaging regions according to characteristics of image processing corresponding to each of the imaging regions, an identification unit configured to identify, after identifying which of the groups the imaging region of the image corresponds to, which of the plurality of the imaging regions included in the corresponding group the imaging region of the image corresponds to, based on the image and the classification information, and a processing unit configured to apply the image processing corresponding to the identified imaging region to the image. | 08-04-2011 |
20110206275 | IMAGE ORIENTATION DETERMINATION DEVICE, IMAGE ORIENTATION DETERMINATION METHOD, AND IMAGE ORIENTATION DETERMINATION PROGRAM - When positive image similarity (degree of training image similarity between input image features and those of positive training image) is higher than a predetermined first threshold, image orientation determination decision section determines input image orientation. When negative image similarity (degree of training image similarity between input image features and those of a negative training image) is not lower than a predetermined second threshold value, the image orientation determination decision section does not determine input image orientation. When the image orientation determination decision section determines the orientation of the input image, image orientation determination section calculates orientation similarity reflecting similarity between input image features and those stored in orientation-specific features storage section. If the calculated orientation similarity satisfies a predetermined condition, the image orientation determination section determines input image orientation according to positive training image orientation related to the image features stored in the orientation-specific features storage section. | 08-25-2011 |
20110216964 | META-CLASSIFIER SYSTEM FOR VIDEO ANALYTICS - A system for meta-classification having a training phase mechanism and an operational phase mechanism. The training phase mechanism may have a detection and tracking module, a classifier section connected to the detection and tracking module, a feature synthesis module connected to the classifier section, a labeling module connected to the feature synthesis module and a training data module connected to the labeling module. The operational phase mechanism may have a detection and tracking module, a classifier section connected to the detection and tracking module, a feature synthesis module connected to the classifier section and a meta-classification module connected to the feature synthesis module and the training module. The training phase mechanism may provide parameters and settings to the operational phase mechanism. | 09-08-2011 |
20110216965 | Image Segmentation Using Reduced Foreground Training Data - Methods of image segmentation using reduced foreground training data are described. In an embodiment, the foreground and background training data for use in segmentation of an image is determined by optimization of a modified energy function. The modified energy function is the energy function used in image segmentation with an additional term comprising a scalar value. The optimization is performed for different values of the scalar to produce multiple initial segmentations and one of these segmentations is selected based on pre-defined criteria. The training data is then used in segmenting the image. In other embodiments further methods are described: one places an ellipse inside the user-defined bounding box to define the background training data and another uses a comparison of properties of neighboring image elements, where one is outside the user-defined bounding box, to reduce the foreground training data. | 09-08-2011 |
20110222759 | INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, AND PROGRAM - An information processing apparatus includes a characteristic amount calculating unit calculating a characteristic amount for each of a plurality of n different image patterns, a specifying unit specifying a best-matching image pattern among the plurality of n image patterns for each of frames forming a learning moving picture and having temporal continuity, a computing unit computing a collocation probability Pij indicating a probability that, for a frame located at a position where a temporal distance to a frame for which a first image pattern Xi is specified among the plurality of n image patterns is within a predetermined threshold τ, a second image pattern Xj is specified among the plurality of n image patterns, and a grouping unit grouping the plurality of n image patterns by using the computed collocation probability Pij. | 09-15-2011 |
20110222760 | METHOD FOR IDENTIFYING MARKED IMAGES BASED AT LEAST IN PART ON FREQUENCY DOMAIN COEFFICIENT DIFFERENCES - Briefly, in accordance with one embodiment, a method of identifying marked images based at least in part on frequency domain coefficient differences is disclosed. | 09-15-2011 |
20110229017 | ANNOTATION ADDITION METHOD, ANNOTATION ADDITION SYSTEM USING THE SAME, AND MACHINE-READABLE MEDIUM - Disclosed are a method and a system for adding annotations into an input medium file. The method comprises a step of creating annotation detection models based on training samples formed by existing media files having annotations; a step of extracting coexistence coefficients of any two annotations based on appearance frequencies of the annotations in the training samples; a step of inputting the input medium file; a step of extracting sense-of-vision features from the input medium file; a step of obtaining initial annotations of the input medium file; a step of acquiring candidate annotations based on the initial annotations and the coexistence coefficients of the annotations in the training samples; and a step of selecting a final annotation set from the candidate annotations based on the sense-of-vision features of the input medium file and the coexistence coefficients by using the annotation detection models. | 09-22-2011 |
20110229018 | CENTRALIZED INFORMATION PROCESSING APPARATUS AND CENTRALIZED INFORMATION PROCESSING SYSTEM - According to one embodiment, a centralized information processing apparatus includes an information acquisition unit configured to acquire image data, sorting destinations arranged in a descending order of scores obtained by character recognition and score information items thereof, and sorting information, a recognition-rate processing unit configured to provide information related to a recognition rate for each sorting destination, a changed parameter value acquisition unit configured to acquire a new parameter value, a simulation executing unit configured to execute a simulation of a character recognition process for the image data by using the changed parameter value, a difference list providing unit configured to form and provide a difference list indicating different content between new sorting information obtained as the simulation result and original sorting information, and a parameter changing unit configured to change the parameter value to the new parameter value. | 09-22-2011 |
20110229019 | Scene Adaptive Brightness/Contrast Enhancement - A method for brightness and contrast enhancement includes computing a luminance histogram of a digital image, computing first distances from the luminance histogram to a plurality of predetermined luminance histograms, estimating first control point values for a global tone mapping curve from predetermined control point values corresponding to a subset of the predetermined luminance histograms selected based on the computed first distances, and interpolating the estimated control point values to determine the global tone mapping curve. The method may also include dividing the digital image into a plurality of image blocks, and enhancing each pixel in the digital image by computing second distances from a pixel in an image block to the centers of neighboring image blocks, and computing an enhanced pixel value based on the computed second distances, predetermined control point values corresponding to the neighboring image blocks, and the global tone mapping curve. | 09-22-2011 |
20110229020 | LEARNING METHOD AND APPARATUS FOR PATTERN RECOGNITION - A method for information processing includes a learning process to generate a tree structured dictionary based on a plurality of patterns including a target object to be recognized. The method includes selecting a plurality of points from an input pattern based on a distribution of a probability that the target object to be recognized is present in the input pattern at each node of a tree structure generated in the learning process, and classifying the input pattern into a branch based on a value of a predetermined function that corresponds to values of the input pattern at selected plurality of points. | 09-22-2011 |
20110235900 | Method for Training Multi-Class Classifiers with Active Selection and Binary Feedback - A multi-class classifier is trained by selecting a query image from a set of active images based on a membership probability determined by the classifier, wherein the active images are unlabeled. A sample image is selected from a set of training image based on the membership probability of the query image, wherein the training images are labeled. The query image and the sample images are displayed to a user on an output device. A response from the user is obtained with an input device, wherein the response is a yes-match or a no-match. The query image with the label of the sample image is added to the training set if the yes-match is obtained, and otherwise repeating the selecting, displaying, and obtaining steps until a predetermined number of no-match is reached to obtain the multi-class classifier. | 09-29-2011 |
20110235901 | METHOD, APPARATUS, AND PROGRAM FOR GENERATING CLASSIFIERS - Classifiers, which are combinations of a plurality of weak classifiers, for discriminating objects included in detection target images by employing features extracted from the detection target images to perform multi class discrimination including a plurality of classes regarding the objects are generated. When the classifiers are generated, branching positions and branching structures of the weak classifiers of the plurality of classes are determined, according to the learning results of the weak classifiers in each of the plurality of classes. | 09-29-2011 |
20110243426 | METHOD, APPARATUS, AND PROGRAM FOR GENERATING CLASSIFIERS - Classifiers, which are combinations of a plurality of weak classifiers, for discriminating objects included in detection target images by employing features extracted from the detection target images to perform multi class discrimination including a plurality of classes regarding the objects are generated. When the classifiers are generated, learning is performed for the weak classifiers of the plurality of classes, sharing only the features. | 10-06-2011 |
20110274345 | ACCURACY OF RECOGNITION BY MEANS OF A COMBINATION OF CLASSIFIERS - In one embodiment, there is provided a method for an Optical Character Recognition (OCR) system. The method comprises: recognizing an input character based on a plurality of classifiers, wherein each classifier generates an output by comparing the input character with a plurality of trained patterns; grouping the plurality of classifiers based on a classifier grouping criterion; and combining the output of each of the plurality of classifiers based on the grouping. | 11-10-2011 |
20110280474 | AUTO CLASSIFYING IMAGES AS "IMAGE NOT AVAILABLE" IMAGES - An image may be accepted from a vendor, and the image may be submitted to an image analysis system. The image analysis system may determine whether the image is a not found image or a true image. The determination may occur in a variety of ways by examining the color and intensity characteristics of an image. After the analysis, a determination is received from the image analysis system of whether the image is a not found image or a true image. | 11-17-2011 |
20110293173 | Object Detection Using Combinations of Relational Features in Images - A classifier for detecting objects in images is constructed from a set of training images. For each training image, features are extracted from a window in the training image, wherein the window contains the object, and then randomly sample coefficients c of the features. N-combinations for each possible set of the coefficients are determined. For each possible combination of the coefficients, a Boolean valued proposition is determined using relational operators to generate a propositional space. Complex hypotheses of a classifier are defined by applying combinatorial functions of the Boolean operators to the propositional space to construct all possible logical propositions in the propositional space. Then, the complex hypotheses of the classifier can be applied to features in a test image to detect whether the test image contains the object. | 12-01-2011 |
20110299764 | METHOD FOR AUTOMATED CATEGORIZATION OF HUMAN FACE IMAGES BASED ON FACIAL TRAITS - A method for automated categorization of human face images based on facial traits, said method comprising a facial trait extracting phase, comprising the steps of: providing a multitude of images comprising human faces, for each image sampling a multitude of points in said image to obtain point sample data, for each sampled point extracting visual features from said point sample data, for each image assigning said visual features to predefined codewords by applying a codebook transform, for each image extracting facial traits by applying a kernel-based learning method's prediction algorithm to said codewords to establish the probability that a facial trait from a predefined set of facial traits is present in said image, and extract said facial trait for said image if said probability is higher than a predefined threshold. | 12-08-2011 |
20110299765 | ROBUST PATTERN RECOGNITION SYSTEM AND METHOD USING SOCRATIC AGENTS - A computer-implemented pattern recognition method, system and program product, the method comprising in one embodiment: creating electronically a linkage between a plurality of models within a classifier module within a pattern recognition system such that any one of said plurality of models may be selected as an active model in a recognition process; creating electronically a null hypothesis between at least one model of said plurality of linked models and at least a second model among said plurality of linked models; accumulating electronically evidence to accept or reject said null hypothesis until sufficient evidence is accumulated to reject said null hypothesis in favor of one of said plurality of linked models or until a stopping criterion is met; and transmitting at least a portion of the electronically accumulated evidence or a summary thereof to accept or reject said null hypothesis to a pattern classifier module. | 12-08-2011 |
20110305384 | INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, AND PROGRAM - An information processing apparatus includes a first generation unit that generates learning images corresponding to a learning moving image, a first synthesis unit that generates a synthesized learning image such that a plurality of the learning images is arranged at a predetermined location and synthesized, a learning unit that computes a feature amount of the generated synthesized learning image, and performs statistical learning using the feature amount to generate a classifier, a second generation unit that generates determination images, a second synthesis unit that generates a synthesized determination image such that a plurality of the determination images is arranged at a predetermined location and synthesized, a feature amount computation unit that computes a feature amount of the generated synthesized determination image, and a determination unit that determines whether or not the determination image corresponds to a predetermined movement. | 12-15-2011 |
20120002868 | METHOD FOR FAST SCENE MATCHING - A method for identifying digital images having matching backgrounds from a collection of digital images, comprising using a processor to perform the steps of: determining a set of one or more feature values for each digital image in the collection of digital images, wherein the set of feature values includes an edge compactness feature value that is an indication of the number of objects in the digital image that are useful for scene matching; determining a subset of the collection of digital images that are good candidates for scene matching by applying a classifier responsive to the determined feature values; applying a scene matching algorithm to the subset of the collection of digital images to identify groups of digital images having matching backgrounds; and storing an indication of the identified groups of digital images having matching backgrounds in a processor-accessible memory. | 01-05-2012 |
20120002869 | System and method for detection of multi-view/multi-pose objects - The present invention provides a computer implemented process for detecting multi-view multi-pose objects. The process comprises training of a classifier for each intra-class exemplar, training of a strong classifier and combining the individual exemplar-based classifiers with a single objective function. This function is optimized using the two nested AdaBoost loops. The first loop is the outer loop that selects discriminative candidate exemplars. The second loop, the inner loop selects the discriminative candidate features on the selected exemplars to compute all weak classifiers for a specific position such as a view/pose. Then all the computed weak classifiers are automatically combined into a final classifier (strong classifier) which is the object to be detected. | 01-05-2012 |
20120033874 | Learning weights of fonts for typed samples in handwritten keyword spotting - A wordspotting system and method are disclosed. The method includes receiving a keyword and, for each of a set of typographical fonts, synthesizing a word image based on the keyword. A keyword model is trained based on the synthesized word images and the respective weights for each of the set of typographical fonts. Using the trained keyword model, handwritten word images of a collection of handwritten word images which match the keyword are identified. The weights allow a large set of fonts to be considered, with the weights indicating the relative relevance of each font for modeling a set of handwritten word images. | 02-09-2012 |
20120039527 | COMPUTER-READABLE MEDIUM STORING LEARNING-MODEL GENERATING PROGRAM, COMPUTER-READABLE MEDIUM STORING IMAGE-IDENTIFICATION-INFORMATION ADDING PROGRAM, LEARNING-MODEL GENERATING APPARATUS, IMAGE-IDENTIFICATION-INFORMATION ADDING APPARATUS, AND IMAGE-IDENTIFICATION-INFORMATION ADDING METHOD - A computer-readable medium storing a learning-model generating program causing a computer to execute a process is provided. The process includes: extracting feature values from an image for learning that is an image whose identification information items are already known, the identification information items representing the content of the image; generating learning models by using binary classifiers, the learning models being models for classifying the feature values and associating the identification information items and the feature values with each other; and optimizing the learning models for each of the identification information items by using a formula to obtain conditional probabilities, the formula being approximated with a sigmoid function, and optimizing parameters of the sigmoid function so that the estimation accuracy of the identification information items is increased. | 02-16-2012 |
20120045120 | INFORMATION PROCESSING APPARATUS AND INFORMATION PROCESSING METHOD - An information processing apparatus that discriminates the orientation of a target includes a calculation unit that calculates a distribution of a difference in feature amount between a plurality of learning patterns each showing the orientation of a target, a determination unit that determines, using a probability distribution obtained from the distribution of differences calculated by the calculation unit, a pixel that is in an input pattern and is to be referred to in order to discriminate the orientation of a target in the input pattern, and a discrimination unit that performs discrimination for obtaining the orientation of the target in the input pattern by comparing a feature amount of the pixel determined by the determination unit and a threshold set in advance. | 02-23-2012 |
20120063673 | METHOD AND APPARATUS TO GENERATE OBJECT DESCRIPTOR USING EXTENDED CURVATURE GABOR FILTER - A method and apparatus to generate an object descriptor using extended curvature gabor filters. The method and apparatus may increase a recognition rate of even a relatively small image with use of an extended number of curvature gabor filters having controllable curvatures and may reduce the amount of calculation required for face recognition by performing the face recognition using only some of the extended curvature gabor filters which have a great effect on the recognition rate. The object descriptor generating method includes extracting gabor features from an input object image by applying a plurality of curvature gabor filters, generated via combination of a plurality of curvatures and a plurality of Gaussian magnitudes, to the object image, and generating an object descriptor for object recognition by projecting the extracted features onto a predetermined base vector. | 03-15-2012 |
20120063674 | PATTERN RECOGNITION APPARATUS AND METHOD THEREFOR CONFIGURED TO RECOGNIZE OBJECT AND ANOTHER LOWER-ORDER OBJECT - In a pattern recognition apparatus, a characteristic amount calculation unit calculates a characteristic amount for recognizing a desired object from a partial image clipped from an input pattern, a likelihood calculation unit calculates a likelihood of an object as a recognition target from the characteristic amount calculated by the characteristic amount calculation unit by referring to an object dictionary, and an object determination unit determines whether the partial image is the object as the recognition target based on the likelihood of the object calculated by the likelihood calculation unit. The likelihood calculation unit calculates the likelihood of the object as the recognition target from the characteristic amount calculated by the characteristic amount calculation unit by referring to a specific object dictionary. The object determination unit determines whether the partial image is a specific object as the recognition target from the likelihood of the object calculated by the likelihood calculation unit. | 03-15-2012 |
20120070073 | SEARCHING DOCUMENT IMAGES - Disclosed is a method of searching a digital image of a document for a predetermined keyword. The method identifies a word in the digital image, the word comprising one or more shapes. A test matrix comprising a difference vector for each character of the word is generated, and a template matrix comprising a difference vector for each shape of the keyword is also generated, wherein a difference vector represents the differences between the visual features of a respective shape and the visual features of a collection of reference shapes. A measure of similarity between the word and the keyword is generated by comparing the test matrix and the template matrix. | 03-22-2012 |
20120070074 | Method and System for Training a Landmark Detector using Multiple Instance Learning - An apparatus and method for training a landmark detector receives training data which includes a plurality of positive training bags, each including a plurality of positively annotated instances, and a plurality of negative training bags, each including at least one negatively annotated instance. Classification function is initialized by training a first weak classifier based on the positive training bags and the negative training bags. All training instances are evaluated using the classification function. For each of a plurality of remaining classifiers, a cost value gradient is calculated based on spatial context information of each instance in each positive bag evaluated by the classification function. A gradient value associated with each of the remaining weak classifiers is calculated based on the cost value gradients, and a weak classifier is selected which has a lowest associated gradient value and given a weighting parameter and added to the classification function. | 03-22-2012 |
20120076401 | IMAGE CLASSIFICATION EMPLOYING IMAGE VECTORS COMPRESSED USING VECTOR QUANTIZATION - Local descriptors are extracted from an image. An image vector is generated having vector elements indicative of parameters of mixture model components of a mixture model representing the extracted local descriptors. The image vector is compressed using a vector quantization algorithm to generate a compressed image vector. Optionally, the compressing comprises splitting the image vector into a plurality of sub-vectors each including at least two vector elements, compressing each sub-vector independently using the vector quantization algorithm, and concatenating the compressed sub-vectors to generate the compressed image vector. Optionally, each sub-vector includes only vector elements indicative of parameters of a single mixture model component, and any sparse sub-vector whose vector elements are indicative of parameters of a mixture model component that does not represent any of the extracted local descriptors is not compressed. | 03-29-2012 |
20120082371 | LABEL EMBEDDING TREES FOR MULTI-CLASS TASKS - Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for label embedding trees for large multi-class tasks. In one aspect, a method includes mapping each image in a plurality of images and each label in a plurality of labels into a multi-dimensional label embedding space. A tree of label predictors is trained with the plurality of mapped images such that an error function is minimized in which the error function counts an error for each mapped image if any of the label predictors at any depth of the tree incorrectly predicts that the mapped image belongs to the label predictor's respective label set. | 04-05-2012 |
20120082372 | AUTOMATIC DOCUMENT IMAGE EXTRACTION AND COMPARISON - Systems and methods are described that extract and match images from a first document with images in other documents. A user controls a threshold on the level of image noise to be ignored and a page range for faster processing of large documents. | 04-05-2012 |
20120087574 | LEARNING DEVICE, LEARNING METHOD, IDENTIFICATION DEVICE, IDENTIFICATION METHOD, AND PROGRAM - Provided is a learning device including: an acquisition section that acquires a plurality of image pairs in which the same subjects appear and a plurality of image pairs in which different subjects appear; a setting section that sets feature points on one image and the other image of each image pair; a selection section that selects a plurality of prescribed feature points, which are set at the same positions of the one image and the other image, so as to thereby select a feature extraction filter for each prescribed feature point; an extraction section that extracts the features of the prescribed feature points of each of the one image and the other image by using the plurality of feature extraction filters; a calculation section that calculates a correlation between the features; and a learning section that learns a same-subject classifier on the basis of the correlation and label information. | 04-12-2012 |
20120087575 | RECOGNIZING HAND POSES AND/OR OBJECT CLASSES - There is a need to provide simple, accurate, fast and computationally inexpensive methods of object and hand pose recognition for many applications. For example, to enable a user to make use of his or her hands to drive an application either displayed on a tablet screen or projected onto a table top. There is also a need to be able to discriminate accurately between events when a user's hand or digit touches such a display from events when a user's hand or digit hovers just above that display. A random decision forest is trained to enable recognition of hand poses and objects and optionally also whether those hand poses are touching or not touching a display surface. The random decision forest uses image features such as appearance, shape and optionally stereo image features. In some cases, the training process is cost aware. The resulting recognition system is operable in real-time. | 04-12-2012 |
20120093396 | DIGITAL IMAGE ANALYSIS UTILIZING MULTIPLE HUMAN LABELS - Systems and methods for implementing a multi-label image recognition framework for classifying digital images are provided. The provided multi-label image recognition framework utilizes an iterative, multiple analysis path approach to model training and image classification tasks. A first iteration of the multi-label image recognition framework generates confidence maps for each label, which are shared by the multiple analysis paths to update the confidence maps in subsequent iterations. The provided multi-label image recognition framework permits model training and image classification tasks to be performed more accurately than conventional single-label image recognition frameworks. | 04-19-2012 |
20120093397 | Method and System for Learning Based Object Detection in Medical Images - Methods and Systems for training a learning based classifier and object detection in medical images is disclosed. In order to train a learning based classifier, positive training samples and negative training samples are generated based on annotated training images. Features for the positive training samples and the negative training samples are extracted. The features include an extended Haar feature set including tip features and corner features. A discriminative classifier is trained based on the extracted features. | 04-19-2012 |
20120093398 | SYSTEM AND METHOD FOR MULTI-AGENT EVENT DETECTION AND RECOGNITION - A method and system for creating a histogram of oriented occurrences (HO2) is disclosed. A plurality of entities in at least one image are detected and tracked. One of the plurality of entities is designated as a reference entity. A local 2-dimensional ground plane coordinate system centered on and oriented with respect to the reference entity is defined. The 2-dimensional ground plane is partitioned into a plurality of non-overlapping bins, the bins forming a histogram, a bin tracking a number of occurrences of an entity class. An occurrence of at least one other entity of the plurality of entities located in the at least one image may be associated with one of the plurality of non-overlapping bins. A number of occurrences of entities of at least one entity class in at least one bin may be into a vector to define an HO2 feature. | 04-19-2012 |
20120099783 | GENERATION AND USAGE OF ATTRACTIVENESS SCORES - A digital image is obtained. A face depicted in the digital image is detected. A set of characteristics is obtained, where the set of characteristics are associated with at least some portion of a face. An attractiveness score is generated based at least in part on the detected face and the set of characteristics. | 04-26-2012 |
20120121170 | OBJECT DETECTION SYSTEM BASED ON A POOL OF ADAPTIVE FEATURES - A method, system and computer program product for detecting presence of an object in an image are disclosed. According to an embodiment, a method for detecting a presence of an object in an image comprises: receiving multiple training image samples; determining a set of adaptive features for each training image sample, the set of adaptive features matching the local structure of each training image sample; integrating the sets of adaptive features of the multiple training image samples to generate an adaptive feature pool; determining a general feature based on the adaptive feature pool; and examining the image using a classifier determined based on the general feature to detect the presence of the object. | 05-17-2012 |
20120128237 | SUPERPIXEL-BOOSTED TOP-DOWN IMAGE RECOGNITION METHODS AND SYSTEMS - Systems and methods for implementing a superpixel boosted top-down image recognition framework are provided. The framework utilizes superpixels comprising contiguous pixel regions sharing similar characteristics. Feature extraction methods described herein provide non-redundant image feature vectors for classification model building. The provided framework differentiates a digitized image into a plurality of superpixels. The digitized image is characterized through image feature extraction methods based on the plurality of superpixels. Image classification models are generated from the extracted image features and ground truth labels and may then be used to classify other digitized images. | 05-24-2012 |
20120128238 | IMAGE PROCESSING DEVICE AND METHOD, DATA PROCESSING DEVICE AND METHOD, PROGRAM, AND RECORDING MEDIUM - An eigenprojection matrix (# | 05-24-2012 |
20120134577 | INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, AND PROGRAM - An information processing apparatus includes: a distinguishing unit which, by using an ensemble classifier, which includes a plurality of weak classifiers outputting weak hypotheses which indicates whether a predetermined subject is shown in an image in response to inputs of a plurality of features extracted from the image, and a plurality of features extracted from an input image, sequentially integrates the weak hypotheses output by the weak classifiers in regard to the plurality of features and distinguishes whether the predetermined subject is shown in the input image based on the integrated value. The weak classifier classifies each of the plurality of features to one of three or more sub-divisions based on threshold values, calculates sum divisions of the sub-divisions of the plurality of features as whole divisions into which the plurality of features is classified, and outputs, as the weak hypothesis, a reliability degree of the whole divisions. | 05-31-2012 |
20120134578 | SYSTEM AND METHOD FOR DETECTING GLOBAL HARMFUL VIDEO - A system for detecting a global harmful video includes: a video determination policy generation unit for determining harmfulness of learning video segments from video learning information to analyze occurrence information of harmful learning video segments, and generating a global harmfulness determination policy based on the occurrence information; and a video determination policy execution unit for determining harmfulness of input video segments from information of an input video to analyze occurrence information of harmful input video segments, and determining whether the input video is harmful or not based on the occurrence information of the harmful input video segments and the global harmfulness determination policy. | 05-31-2012 |
20120134579 | IMAGE PROCESSING DEVICE AND METHOD, DATA PROCESSING DEVICE AND METHOD, PROGRAM, AND RECORDING MEDIUM - In an image processing device and method, program, and recording medium of the present invention, high frequency components of a low quality image and a high quality image included in a studying image set are extracted, and an eigenprojection matrix and a projection core tensor of the high frequency components are generated in a studying step. In a restoration step, a first sub-core tensor and a second sub-core tensor are generated based on the eigenprojection matrix and the projection core tensor of the high frequency components, and a tensor projection process is applied to the high frequency components of an input image to generate a high quality image of the high frequency components. The high quality image of the high frequency components is added to an enlarged image obtained by enlarging the input image to the same size as an output image. | 05-31-2012 |
20120141017 | REDUCING FALSE DETECTION RATE USING LOCAL PATTERN BASED POST-FILTER - A training set for a post-filter classifier is created from the output of a face detector. The face detector can be a Viola Jones face detector. Face detectors produce false positives and true positives. The regions in the training set are labeled so that false positives are labeled negative and true positives are labeled positive. The labeled training set is used to train a post-filter classifier. The post-filter classifier can be an SVM (Support Vector Machine). The trained face detection classifier is placed at the end of a face detection pipeline comprising a face detector, one or more feature extractors and the trained post-filter classifier. The post-filter reduces the number of false positives in the face detector output while keeping the number of true positives almost unchanged using features different from the Haar features used by the face detector. | 06-07-2012 |
20120141018 | L1-Optimized AAM Alignment - An Active Appearance Model, AAM, uses an L | 06-07-2012 |
20120141019 | REGION DESCRIPTION AND MODELING FOR IMAGE SUBSCENE RECOGNITION - A method and apparatus is described here that categorizes images by extracting regions and describing the regions with a 16-dimensional subscene feature vector, which is a concatenation of color, texture, and spatial feature vectors. By comparing the spatial feature vectors in images with similarly-obtained feature vectors in a Gaussian mixture based model pool (obtained in a subscene modeling phase), the images may be categorized (in a subscene recognition phase) with probabilities relating to each region or subscene. Higher probabilities are likelier correlations. The device may be a single or multiple core CPU, or parallelized vector processor for characterizing many images. The images may be photographs, videos, or video stills, without restriction. When used real-time, the method may be used for visual searching or sorting. | 06-07-2012 |
20120141020 | IMAGE CLASSIFICATION - Images are classified as photos (e.g., natural photographs) or graphics (e.g., cartoons, synthetically generated images), such that when searched (online) with a filter, an image database returns images corresponding to the filter criteria (e.g., either photos or graphics will be returned). A set of image statistics pertaining to various visual cues (e.g., color, texture, shape) are identified in classifying the images. These image statistics, combined with pre-tagged image metadata defining an image as either a graphic or a photo, may be used to train a boosting decision tree. The trained boosting decision tree may be used to classify additional images as graphics or photos based on image statistics determined for the additional images. | 06-07-2012 |
20120148148 | METHOD FOR DETECTING FIRE-FLAME USING FUZZY FINITE AUTOMATA - A method for detecting a fire flame using fuzzy finite automata is provided. The fire-flame detection method comprises (1) acquiring an image required for the detection of fire-flame, (2) dividing the image into a number of blocks, (3) extracting a fire-flame candidate block using a brightness distortion of a pixel in the block, (4) detecting a fire-flame candidate region from the fire-flame block using a color probability model, and (5) determining whether the fire-flame candidate region corresponds to a fire-flame via fuzzy finite automata. The fire-flame detection method can detect fire-flames in a variety of fire images with relatively high precision, by establishing a probability model using the brightness distortion and wavelet energy in fire-flame regions with continuous and irregular fluctuation patterns and using the upward motion, and applying the model to fuzzy finite automata. | 06-14-2012 |
20120155751 | OBJECT RECOGNITION APPARATUS, OBJECT RECOGNITION METHOD, LEARNING APPARATUS, LEARNING METHOD, STORAGE MEDIUM AND INFORMATION PROCESSING SYSTEM - A learning method of detectors used to detect a target object, comprises: a selection step of selecting a plurality of specific regions from a given three-dimensional model of the target object; a learning step of learning detectors used to detect the specific regions selected in the selection step; an evaluation step of executing recognition processing of positions and orientations of predetermined regions of the plurality of specific regions by the detectors learned in the learning step; and a normalization step of setting vote weights for outputs of the detectors according to recognition accuracies of results of the recognition processing in the evaluation step. | 06-21-2012 |
20120163706 | SHAPE DESCRIPTION AND MODELING FOR IMAGE SUBSCENE RECOGNITION - A method and apparatus is described here that categorizes images by extracting a subscene and describing the subscene with a top level feature vector and a division feature vector, which are descriptions of edge gradient classifications within rectangular bounding boxes. By filtering subscene feature vectors in images with a Gaussian mixture based model pool (obtained in a subscene modeling phase), the images may be categorized (in an subscene recognition phase) with probabilities relating to each subscene. Higher probabilities are likelier correlations. The device may be a single or multiple core CPU, or parallelized vector processor for characterizing many images. The images may be photographs, videos, or video stills, without restriction. When used real-time, the method may be used for visual searching or sorting. | 06-28-2012 |
20120163707 | MATCHING TEXT TO IMAGES - Text in web pages or other text documents may be classified based on the images or other objects within the webpage. A system for identifying and classifying text related to an object may identify one or more web pages containing the image or similar images, determine topics from the text of the document, and develop a set of training phrases for a classifier. The classifier may be trained and then used to analyze the text in the documents. The training set may include both positive examples and negative examples of text taken from the set of documents. A positive example may include captions or other elements directly associated with the object, while negative examples may include text taken from the documents, but from a large distance from the object. In some cases, the system may iterate on the classification process to refine the results. | 06-28-2012 |
20120163708 | APPARATUS FOR AND METHOD OF GENERATING CLASSIFIER FOR DETECTING SPECIFIC OBJECT IN IMAGE - There provides an apparatus for and a method of generating a classifier for detecting a specific object in an image. The apparatus for generating a classifier for detecting a specific object in an image includes: a region dividing section for dividing, from a sample image, at least one square region having a side length equal to or shorter than the length of shorter side of the sample image; a feature extracting section for extracting an image feature from at least a part of the square regions divided by the region dividing section; and a training section for performing training based on the extracted image feature to generate a classifier. By using the apparatus for and method of generating the classifier, it becomes possible to make full use of recognizable regions of objects to be recognized with variable aspect ratios and improve speed and accuracy for recognizing in complex backgrounds. | 06-28-2012 |
20120170834 | Determining the Uniqueness of a Model for Machine Vision - Described are methods and apparatuses, including computer program products, for determining model uniqueness with a quality metric of a model of an object in a machine vision application. Determining uniqueness involves receiving a training image, generating a model of an object based on the training image, generating a modified training image based on the training image, determining a set of poses that represent possible instances of the model in the modified training image, and computing a quality metric of the model based on an evaluation of the set of poses with respect to the modified training image. | 07-05-2012 |
20120170835 | Determining the Uniqueness of a Model for Machine Vision - Described are methods and apparatuses, including computer program products, for determining model uniqueness with a quality metric of a model of an object in a machine vision application. Determining uniqueness involves receiving a training image and a first set of model parameters, generating a first model of an object, generating a second model of the object based on the training image and a second set of model parameters modified from the first set of model parameters, determining a set of poses that represent possible instances of the second model in the training image, and computing a quality metric of the first model based on an evaluation of the set of poses with respect to the training image. | 07-05-2012 |
20120183206 | INTERACTIVE CONCEPT LEARNING IN IMAGE SEARCH - An interactive concept learning image search technique that allows end-users to quickly create their own rules for re-ranking images based on the image characteristics of the images. The image characteristics can include visual characteristics as well as semantic features or characteristics, or may include a combination of both. End-users can then rank or re-rank any current or future image search results according to their rule or rules. End-users provide examples of images each rule should match and examples of images the rule should reject. The technique learns the common image characteristics of the examples, and any current or future image search results can then be ranked or re-ranked according to the learned rules. | 07-19-2012 |
20120189192 | Imaging Method and Apparatus with Optimized Grayscale Value Window Determination - A self-learning imaging method is particularly suited radiation imaging, such as for mammography. A plurality of training data sets are displayed on a display apparatus. A grayscale value setting is selected for each training data set. A feature set with at least one feature is assigned to each training data set. The feature set and the grayscale value setting are stored for each training data set. The grayscale value setting of an examination data set is selected according to the feature sets and the grayscale value setting of the training data sets. | 07-26-2012 |
20120189193 | DETECTION OF OBJECTS REPRESENTED IN IMAGES - The invention relates to computer vision, in particular detection and classification of objects captured in a video stream of images. The invention provides a memory efficient method of storing images that have been pre-processed for use in object detection. The method is based on using histograms of orientation. The invention also includes methods for training and using weak classifiers that use this pre-processing of images. A first weak classifier uses the total count of two orientation values in a histogram as an index to a two dimensional confidence table to determine a confidence value. The second weak classifier projects one or more total counts of orientation values in a histogram into a scalar value that is then used in a one dimensional confidence map to determine a confidence value. Aspects of the invention include methods, computer systems and software. | 07-26-2012 |
20120195495 | Hierarchical Tree AAM - An active appearance model is built by arranging the training images in its training library into a hierarchical tree with the training images at each parent node being divided into two child nodes according to similarities in characteristic features. The number of node levels is such that the number of training images associated with each leaf node is smaller than a predefined maximum. A separate AAM, one per leaf node, is constructed using each leaf node's corresponding training images. In operation, starting at the root node, a test image is compared with each parent node's two child nodes and follows a node-path of model images that most closely matches the test image. The test image is submitted to an AAM selected for being associated with the leaf node at which the test image rests. The selected AAM's output aligned image may be resubmitted to the hierarchical tree if sufficient alignment is not achieved. | 08-02-2012 |
20120213426 | Method for Implementing a High-Level Image Representation for Image Analysis - Robust low-level image features have been proven to be effective representations for a variety of visual recognition tasks such as object recognition and scene classification; but pixels, or even local image patches, carry little semantic meanings. For high-level visual tasks, such low-level image representations are potentially not enough. The present invention provides a high-level image representation where an image is represented as a scale-invariant response map of a large number of pre-trained generic object detectors, blind to the testing dataset or visual task. Leveraging on this representation, superior performances on high-level visual recognition tasks are achieved with relatively classifiers such as logistic regression and linear SVM classifiers. | 08-23-2012 |
20120213427 | IMAGE PROCESSING APPARATUS AND IMAGE PROCESSING METHOD - Disclosed are an image processing apparatus and an image processing method. The image processing apparatus comprises a matching degree calculation unit configured to calculate respective matching degrees between an image waiting for processing and plural training images whose Kansei scores are pre-designated; and a Kansei score calculation unit configured to extract, from the plural training images, a predetermined number of training images corresponding to the maximum matching degree, and then based on the Kansei scores of the extracted training images with regard to a selected Kansei type, calculate a Kansei score of the image waiting for processing. | 08-23-2012 |
20120213428 | TRAINING DEVICE, TRAINING SYSTEM AND METHOD - A training device comprises a first regenerating unit regenerates at least one of an image and a voice for training during the training courses which lead the user to train the operation of an input device, an operation accepting unit accepts the user operation for at least one of the image and the voice for training from a simulated user interface which simulates a user interface of the input device during training, a second regenerating unit regenerates at least one of the image and the voice for training when the training is ended, and a normal operation instructing unit instructs a normal operation to the user by outputting at least one of the image and the voice of the normal operation of the user, which show at least one of the image and the voice for training, which is synchronous with the regeneration of the second regenerating unit. | 08-23-2012 |
20120219209 | Image Labeling with Global Parameters - Image labeling with global parameters is described. In an embodiment a pose estimation system executes automatic body part labeling. For example, the system may compute joint recognition or body part segmentation for a gaming application. In another example, the system may compute organ labels for a medical imaging application. In an example, at least one global parameter, for example body height is computed for each of the images to be labeled. In an example, the global parameter is used to modify an image labeling process. For example the global parameter may be used to modify the input image to a canonical scale. In another example, the global parameter may be used to adaptively modify previously stored parameters of the image labeling process. In an example, the previously stored parameters may be computed from a reduced set of training data. | 08-30-2012 |
20120219210 | Multi-Scale, Perspective Context, and Cascade Features for Object Detection - Systems and methods for object detection are presented herein. Embodiments of the present invention utilizing a cascade feature, one or more features at different scales, one or more multi-scale features in combination with a perspective feature, or combinations thereof to detect an object of interest in an input image. In embodiments, the various features are used to train classifiers. In embodiments, the trained classifiers are used in detecting an object of interest in one or more input images. | 08-30-2012 |
20120219211 | Contextual Boost for Object Detection - Aspects of the present invention includes systems and methods for generating detection models that consider contextual information of an image patch and for using detection models that consider contextual information. In embodiments, a multi-scale image context descriptor is generated to represent the contextual cues in multiple parameters, such as spatial, scaling, and color spaces. In embodiments, a classification context is defined using the contextual features and is used in a contextual boost classification scheme. In embodiments, the contextual boost propagates contextual cues to larger coverage through iterations to improve the detection accuracy. | 08-30-2012 |
20120219212 | FEATURE CONVERSION DEVICE, SIMILAR INFORMATION SEARCH APPARATUS PROVIDED THEREWITH, CODING PARAMETER GENERATION METHOD, AND COMPUTER PROGRAM - A bit code converter transforms a learning feature vector using a transformation matrix updated by a transformation matrix update unit, and converts the transformed learning feature vector into a bit code. When the transformation matrix update unit substitutes a substitution candidate for an element of the transformation matrix, a cost function calculator fixes the substitution candidate that minimizes a cost function as the element. The transformation matrix update unit selects the element while sequentially changing the elements, and the cost function calculator fixes the selected element every time the transformation matrix update unit selects the element, thereby finally fixing the optimum transformation matrix. A substitution candidate specifying unit specifies the substitution candidate such that a speed of transformation processing that the bit code converter performs using the transformation matrix using the transformation matrix is enhanced based on a constraint condition stored in a constraint condition storage unit. | 08-30-2012 |
20120219213 | Embedded Optical Flow Features - Aspects of the present invention include systems and methods for generating an optical flow-based feature. In embodiments, to extract an optical flow feature, the optical flow at sparse interest points is obtained, and Locality-constrained Linear Coding (LLC) is applied to the sparse interest points to embed each flow into a higher-dimensional code. In embodiments, for an image frame, the multiple codes are combined together using a weighted pooling that is related to the distribution of the optical flows in the image frame. In embodiments, the feature may be used in training models to detect actions, in trained models for action detection, or both. | 08-30-2012 |
20120224764 | METHOD FOR COLOR RECOGNITION - A color recognition method is provided, including a color learning phase and a color recognition phase. The color learning phase further includes the steps of: preparing a plurality of color images printed on at least a sheet of paper; for each color image of the plurality of color images, obtaining the digital signals representing the red, green and blue components of the color; associating the digital signals with an audio file related to the color image; and storing the digital signals with associated audio file to a database in the storage device. The color recognition phase further includes the steps of: obtaining the digital signals representing the red, green and blue components of an color image printed on a sheet of paper; comparing the digital signals representing the color image to a pre-stored database in a storage device; if a match is found, playing the associated audio file. | 09-06-2012 |
20120224765 | TEXT REGION DETECTION SYSTEM AND METHOD - A method for detecting a text region in an image is disclosed. The method includes detecting a candidate text region from an input image. A set of oriented gradient images is generated from the candidate text region, and one or more detection window images of the candidate text region are captured. A sum of oriented gradients is then calculated for a region in one of the oriented gradient images. It is classified whether each detection window image contains text by comparing the associated sum of oriented gradients and a threshold. Based on the classifications of the detection window images, it is determined whether the candidate text region is a true text region. | 09-06-2012 |
20120237116 | Identifying Text Pixels in Scanned Images - A processor and method make use of multiple weak classifiers to construct a single strong classifier to identify regions that contain text within an input image document. The weak classifiers are grouped by their computing cost from low to median to high, and each weak classifier is assigned a weight value based on its ability to accurately identify text regions. A level 1 classifier is constructed by selecting weak classifiers from the low group, a level 2 classifier is constructed by selecting weak classifiers from the low and median groups, and a level 3 classifier is constructed by selecting weak classifiers from the low, median and high groups. Regions that the level 1 classifier identifies as containing text are submitted to the level 2 classifier, and regions that the level 2 classifier identifies as containing text are submitted to the level 3 classifier. | 09-20-2012 |
20120237117 | OPTIMAL GRADIENT PURSUIT FOR IMAGE ALIGNMENT - A method for image alignment is disclosed. In one embodiment, the method includes acquiring a facial image of a person and using a discriminative face alignment model to fit a generic facial mesh to the facial image to facilitate locating of facial features. The discriminative face alignment model may include a generative shape model component and a discriminative appearance model component. Further, the discriminative appearance model component may have been trained to estimate a score function that minimizes the angle between a gradient direction and a vector pointing toward a ground-truth shape parameter. Additional methods, systems, and articles of manufacture are also disclosed. | 09-20-2012 |
20120243778 | IMAGE RECOGNIZING APPARATUS, METHOD FOR RECOGNIZING IMAGE AND NON-TRANSITORY COMPUTER READABLE MEDIUM - An image recognizing apparatus includes a dictionary memory, a block determining module and a recognizing module. The dictionary memory stores dictionary data. The block determining module determines that a target block comprising a target pixel to be processed of a plurality of pixels in image data is a shared block to which the dictionary data is used or a mirror block to which the dictionary data to the shared block is used, based on a position of the target block. The recognizing module uses common dictionary data for the shared block and the mirror block, and recognizes a characteristic portion of the image expressed by the image data. | 09-27-2012 |
20120243779 | RECOGNITION DEVICE, RECOGNITION METHOD, AND COMPUTER PROGRAM PRODUCT - According to an embodiment, a recognition device includes a generation unit to select, plural times, groups each including learning samples from a storage unit, learn a classification metric for classifying the groups selected in each selection, and generate an evaluation metric including the classification metrics; a transformation unit to transform a first feature value of an image including an object into a second feature value using the evaluation metric; a calculation unit to calculate similarities of the object to categories in a table using the second feature value and reference feature values; and a registration unit to register the second feature value as the reference feature value in the table associated with the category of the object and register the first feature value as the learning sample belonging to the category of the object in the storage unit. The generation unit performs the generation again. | 09-27-2012 |
20120250982 | IMAGE PROCESSING APPARATUS, IMAGE PROCESSING METHOD, PROGRAM, AND RECORDING MEDIUM - An image processing apparatus includes: an image feature outputting unit that outputs each of image features in correspondence with a time of the frame; a foreground estimating unit that estimates a foreground image at a time s by executing a view transform as a geometric transform on a foreground view model and outputs an estimated foreground view; a background estimating unit that estimates a background image at the time s by executing a view transform as a geometric transform on a background view model and outputs an estimated background view; a synthesized view generating unit that generates a synthesized view by synthesizing the estimated foreground and background views; a foreground learning unit that learns the foreground view model based on an evaluation value; and a background learning unit that learns the background view model based on the evaluation value by updating the parameter of the foreground view model. | 10-04-2012 |
20120250983 | OBJECT DETECTING APPARATUS AND METHOD - An object detecting apparatus and method is disclosed. An object detecting apparatus comprises: a detection classifier, configured to detect an object in an input image to obtain one or more candidate objects; a verifying classifier, configured to verify each candidate object by using verifying features from an image block corresponding to the each candidate object; and an on-line learning device, configured to train and optimize the detection classifier by using image blocks corresponding to the candidate objects as on-line samples, based on verifying results of the candidate objects obtained by the verifying classifier. | 10-04-2012 |
20120257819 | Vision-Based Object Detection by Part-Based Feature Synthesis - A method is provided for training and using an object classifier to identify a class object from a captured image. A plurality of still images is obtained from training data and a feature generation technique is applied to the plurality of still images for identifying candidate features from each respective image. A subset of features is selected from the candidate features using a similarity comparison technique. Identifying candidate features and selecting a subset of features is iteratively repeated a predetermined number of times for generating a trained object classifier. An image is captured from an image capture device. Features are classified in the captured image using the trained object classifier. A determination is made whether the image contains a class object based on the trained object classifier associating an identified feature in the image with the class object. | 10-11-2012 |
20120257820 | IMAGE ANALYSIS TOOLS - A master image can be generated based upon evaluation of virtual machine images. The master image includes single instances of data segments that are shared across virtual machine images within a virtual machine environment. The master image can be further be constructed as a function of a peer pressure technique that includes data segments common to a majority of the virtual machine images within the master image. The data segments included within the master image can further be defined by prioritizing data within virtual machine images as well as identifying influential data with a peer pressure technique. | 10-11-2012 |
20120269425 | PREDICTING THE AESTHETIC VALUE OF AN IMAGE - A system and method for determining the aesthetic quality of an image are disclosed. The method includes extracting a set of local features from the image, such as gradient and/or color features and generating an image representation which describes the distribution of the local features. A classifier system is used for determining an aesthetic quality of the image based on the computed image representation. | 10-25-2012 |
20120269426 | FEATURE SELECTION METHOD AND APPARATUS, AND PATTERN DISCRIMINATION METHOD AND APPARATUS - A feature selection apparatus, which selects features to be used to discriminate an object by a discriminator using learning data including the object, extracts a plurality of partial data from the learning data, and obtains discrimination values obtained by controlling the discriminator to process the plurality of extracted partial data as features of the plurality of partial data. The feature selection apparatus evaluates the obtained features based on discrimination degrees on a discrimination space defined by the discriminator, and selects features to be used to discriminate the object from a plurality of features obtained in association with the plurality of partial data based on an evaluation result. | 10-25-2012 |
20120275691 | COEFFICIENT LEARNING DEVICE AND METHOD, IMAGE PROCESSING DEVICE AND METHOD, PROGRAM, AND RECORDING MEDIUM - A feature-quantity extraction unit extracts a feature quantity of a target pixel of a student image. The target pixel is classified into a predetermined class. Natural-image processing of the target pixel is performed. Artificial-image processing of the target pixel is performed. A sample of a normal equation is generated using a pixel value of the target pixel subjected to the natural-image processing, a pixel value of the target pixel subjected to the artificial-image processing, a pixel value of a target pixel of a teacher image, and a predetermined mixing coefficient for each class. The mixing coefficient is calculated on the basis of a plurality of generated samples. | 11-01-2012 |
20120275692 | RECOGNITION APPARATUS, RECOGNITION METHOD, AND STORAGE MEDIUM - A recognition apparatus includes a calculation unit configured to calculate likelihood of each feature quantity based on the weighted distribution of the feature quantity extracted from a plurality of learning images, a correction unit configured, if a ratio of a learning image to a specific feature quantity is equal to or smaller than a predetermined ratio and a weight for the specific feature quantity is greater than a predetermined value, to correct the value of likelihood of the specific feature quantity to lower the value based on the distribution, a setting unit configured to set the likelihood corrected by the correction unit in association with a feature quantity, and a discrimination unit to extract a feature quantity from an input image and discriminate whether the input image includes a predetermined object based on the likelihood associated with the feature quantity. | 11-01-2012 |
20120275693 | METHOD FOR IDENTIFYING MARKED CONTENT - Briefly, in accordance with one embodiment, a method of identifying marked content is described. | 11-01-2012 |
20120281907 | REAL-TIME 3D POINT CLOUD OBSTACLE DISCRIMINATOR APPARATUS AND ASSOCIATED METHODOLOGY FOR TRAINING A CLASSIFIER VIA BOOTSTRAPPING - Training a strong classifier by classifying point cloud data with a first classifier, inferring a first terrain map from the classified point cloud data, reclassifying the point cloud data with the first classifier based on the first terrain map, and training a second classifier based on the point cloud data reclassified with the first classifier based on the terrain map. The point cloud data is then classified with the second classifier, and the procedure followed with the first classifier is iteratively repeated until a strong classifier is determined. A strong classifier is determined when a probability of a terrain map matching a given terrain for the strong classifier is approximately equal to a probability of a terrain map matching the given terrain for a prior trained classifier. | 11-08-2012 |
20120281908 | INTELLIGENT AIRFOIL COMPONENT SURFACE IMAGING INSPECTION - A method for inspecting surfaces including acquiring a surface image from a surface of a component; providing an image registration for the surface image; inspecting the component in response to the image registration to produce an input data set; creating an output data set in response to the input data set utilizing a fuzzy logic algorithm; and identifying a surface feature in response to the surface image and the output data set where acquiring the surface image further includes generating a radiation media; directing the radiation media at the component; detecting a responding radiation media in response to the directed radiation media and the component; creating the surface image in response to detecting the responding radiation media; and adjusting the generation of the radiation media in response to the surface image and a standard image. | 11-08-2012 |
20120281909 | LEARNING DEVICE, IDENTIFICATION DEVICE, LEARNING IDENTIFICATION SYSTEM AND LEARNING IDENTIFICATION DEVICE - A learning device includes a gradient feature extraction unit which extracts a gradient feature amount including a gradient direction at each coordinate and a gradient intensity value thereof based on an amount of variation between luminance at each coordinate of an inputted learning target pattern and luminance at a periphery thereof, a sum difference feature extraction unit which calculates a predetermined sum difference feature amount by adding the gradient intensity values according to the gradient directions included in a predetermined gradient range indicating a range of the predetermined gradient direction based on the extracted gradient feature amount and subtracting the gradient intensity values according to the gradient directions included in the other gradient range adjacent to the predetermined gradient range from the calculated sum, and a learning unit which acquires a learning parameter at each coordinate. | 11-08-2012 |
20120281910 | DETECTING FACIAL SIMILARITY BASED ON HUMAN PERCEPTION OF FACIAL SIMILARITY - Similar faces may be determined within images based on human perception of facial similarity. The user may provide an image including a query face to which the user wishes to find faces that are similar. Similar faces may be determined based on similarity information. Similarity information may be generated from information related to a human perception of facial similarity. Images that include faces determined to be similar, based on the similarity information, may be provided to the user as search result images. The user then may provide feedback to indicate the user's perception of similarity between the query face and the search result images. | 11-08-2012 |
20120288186 | SYNTHESIZING TRAINING SAMPLES FOR OBJECT RECOGNITION - An enhanced training sample set containing new synthesized training images that are artificially generated from an original training sample set is provided to satisfactorily increase the accuracy of an object recognition system. The original sample set is artificially augmented by introducing one or more variations to the original images with little to no human input. There are a large number of possible variations that can be introduced to the original images, such as varying the image's position, orientation, and/or appearance and varying an object's context, scale, and/or rotation. Because there are computational constraints on the amount of training samples that can be processed by object recognition systems, one or more variations that will lead to a satisfactory increase in the accuracy of the object recognition performance are identified and introduced to the original images. | 11-15-2012 |
20120288187 | ADDITION RATIO LEARNING APPARATUS AND METHOD, IMAGE PROCESSING APPARATUS AND METHOD, PROGRAM, AND RECORDING MEDIUM - There is provided an addition ratio learning apparatus including a noise adding unit that adds noises to data of an image input as a teacher image, a motion compensating unit that sets an image where time addition noise reduction processing is executed as an NR screen and performs motion compensation with respect to the NR screen, a differential feature amount calculating unit that sets an image as an input screen and calculates a differential feature amount, a circulation history specifying unit that counts a circulation history in the time addition noise reduction processing and specifies the circulation history, an addition ratio computing unit that computes an addition ratio on the basis of pixel values, and a time adding unit that performs multiplication by a coefficient determined according to the computed addition ratio to perform weighted addition and executes the time addition noise reduction processing with respect to the input screen. | 11-15-2012 |
20120294514 | TECHNIQUES TO ENABLE AUTOMATED WORKFLOWS FOR THE CREATION OF USER-CUSTOMIZED PHOTOBOOKS - A system and method for generating a photobook are provided. The method includes receiving a set of images and automatically selecting a subset of the images as candidates for inclusion in a photobook. At least one design element of a design template for the photobook is automatically selected, based on information extracted from at least one of the images in the subset. Placeholders of the design template are automatically filled with images drawn from the subset to form at least one page of a multipage photobook. The exemplary system and method address some of the problems of photobook creation, thorough combining automatic methods for selecting, cropping, and placing photographs into a photo album template, which the user can then post-edit, if desired. This can greatly reduce the time required to create a photobook and thus encourage users to print photo albums. | 11-22-2012 |
20120294515 | IMAGE PROCESSING APPARATUS AND IMAGE PROCESSING METHOD, LEARNING APPARATUS AND LEARNING METHOD, PROGRAM, AND RECORDING MEDIUM - A predictive signal processing unit calculates a pixel value of a luminance component of a pixel of interest by a calculation of a predictive coefficient for a luminance component and a luminance prediction tap. A predictive signal processing unit calculates a pixel value of a chrominance component of a pixel of interest by a calculation of a predictive coefficient for a chrominance component which is higher in noise reduction effect than the predictive coefficient for the luminance component and a chrominance prediction tap. For example, the present technology can be applied to an image processing apparatus. | 11-22-2012 |
20120301014 | LEARNING TO RANK LOCAL INTEREST POINTS - Tools and techniques for learning to rank local interest points from images using a data-driven scale-invariant feature transform (SIFT) approach termed “Rank-SIFT” are described herein. Rank-SIFT provides a flexible framework to select stable local interest points using supervised learning. A Rank-SIFT application detects interest points, learns differential features, and implements ranking model training in the Gaussian scale space (GSS). In various implementations a stability score is calculated for ranking the local interest points by extracting features from the GSS and characterizing the local interest points based on the features being extracted from the GSS across images containing the same visual objects. | 11-29-2012 |
20120301015 | IMAGE IDENTIFICATION DEVICE, IMAGE IDENTIFICATION METHOD AND RECORDING MEDIUM - The invention provides an image identification device uses a separating plane to classify block images into the categories. The image identification device includes a target image input unit inputting a target image, a block image generation unit generates block images, a feature quantity computing unit computes feature quantities of the block images, and a category determination unit determines whether the block images are classified into the categories or not. The feature quantity computing unit uses local feature quantities of the block images and a global feature quantity of the target image as a whole, and also in a feature quantity space using features of the block images as coordinate axes, uses coordinate positions of feature quantity vectors optional areas in the feature quantity space to count the block images and causes the global feature quantity to include the number of the block images thus counted. | 11-29-2012 |
20120308123 | APPARATUS AND METHOD FOR ESTIMATING THE NUMBER OF OBJECTS INCLUDED IN AN IMAGE - An apparatus and method for estimating the number of objects in an input image are disclosed. The apparatus includes: a learning unit that calculates counted values of a linear regression function by learning an arbitrary image; a separation unit that separates a foreground region and a background region of the input image; an extraction unit that searches for features that require an amount of calculation that is below a particular threshold from features having high correlation with each other feature and extracts the features from the separated foreground region; and an estimation unit that estimates the number of objects in the foreground region as a dependent variable by allocating the counted values of the linear regression function that are calculated by the learning unit and the features that are extracted by the extraction unit as independent variables of a linear regression function. | 12-06-2012 |
20120308124 | Method and System For Localizing Parts of an Object in an Image For Computer Vision Applications - A method is provided for localizing parts of an object in an image by training local detectors using labeled image exemplars with fiducial points corresponding to parts within the image. Each local detector generates a detector score corresponding to the likelihood that a desired part is located at a given location within the image exemplar. A non-parametric global model of the locations of the fiducial points is generated for each of at least a portion of the image exemplars. An input image is analyzed using the trained local detectors, and a Bayesian objective function is derived for the input image from the non-parametric model and detector scores. The Bayesian objective function is optimized using a consensus of global models, and an output is generated with locations of the fiducial points labeled within the object in the image. | 12-06-2012 |
20120314938 | Image Type Classifier For Improved Remote Presentation Session Compression - An invention is disclosed for classifying a graphic—e.g. as text or non-text. In embodiments, machine learning is used to generate a solution for classifying graphics of a graphic based on providing the machine learning system a plurality of graphics that are already classified. The way to determine a classification is then used by a remote presentation session server to classify tiles of frames to be transmitted to a client in a remote presentation session. The server encodes the tiles based on their classifications and transmits the encoded tiles to the client. | 12-13-2012 |
20120314939 | RECOGNIZING APPARATUS AND METHOD, PROGRAM, AND RECORDING MEDIUM - A predetermined feature point obtained from an input image is extracted. An image that indicates a locus specifying a predetermined graphic included in the input image and corresponds to a feature point is acquired using a Hough transform. A recognition target object is detected from an input image, based on a plurality of feature quantities, using an identifier generated by statistical learning using the plurality of feature quantities obtained from a locus image obtained based on a learning image including the recognition target object and a locus image obtained based on a learning image including no recognition target object. | 12-13-2012 |
20120314940 | IMAGE RECOGNITION DEVICE AND METHOD OF RECOGNIZING IMAGE THEREOF - An image recognition device in accordance with the inventive concept may include an input vector extraction part extracting an input vector from an input image; a compression vector conversion part converting the input vector into a compression vector using a projection vector; a training parameter generation part receiving a training vector to generate a training parameter using a projection vector obtained through a folding operation of the training vector; and an image classification part classifying the compression vector using the training vector to output image recognition data. | 12-13-2012 |
20120321174 | Image Processing Using Random Forest Classifiers - A method of performing image retrieval includes training a random forest RF classifier based on low-level features of training images and a high-level feature, using similarity values generated by the RF classifier to determine a subset of the training images that are most similar to one another, and classifying input images for the high-level feature using the RF classifier and the determined subset of images. | 12-20-2012 |
20120321175 | LOCATION-AIDED RECOGNITION - A mobile device having the capability of performing real-time location recognition with assistance from a server is provided. The approximate geophysical location of the mobile device is uploaded to the server. Based on the mobile device's approximate geophysical location, the server responds by sending the mobile device a message comprising a classifier and a set of feature descriptors. This can occur before an image is captured for visual querying. The classifier and feature descriptors are computed during an offline training stage using techniques to minimize computation at query time. The classifier and feature descriptors are used to perform visual recognition in real-time by performing the classification on the mobile device itself. | 12-20-2012 |
20120321176 | METHODS AND APPARATUSES FOR FACILITATING OBJECT RECOGNITION - Methods and apparatuses are provided for facilitating object recognition. A method may include accessing data for a first object and data for a second object. The method may additionally include comparing the first and second objects based at least in part upon a reference set and training results generated based at least in part upon the reference set and training data. The method may further include determining whether the first object and the second object are the same object based at least in part upon the comparison. Corresponding apparatuses are also provided. | 12-20-2012 |
20120328184 | OPTICALLY CHARACTERIZING OBJECTS - Systems and methods are provided for optically characterizing an object. A method includes querying an image search engine for the object; extracting image features from multiple images returned by the search engine in response to the query; clustering the image features extracted from the images returned by the search engine according to similarities in optical characteristics of the image features; and determining a set of image features most representative of the object based on the clustering. | 12-27-2012 |
20130004061 | IMAGE PROCESSING DEVICE, IMAGE PROCESSING PROGRAM, AND METHOD FOR GENERATING IMAGE - An image processing device includes a texture component up-sampling portion for up-sampling a texture component of an input image and a component mixing portion for mixing an up-sampled structure component of the input image and the up-sampled texture component obtained by the texture component up-sampling portion, wherein the texture component up-sampling portion up-samples the texture component by means of a learning-based method using a reference image. | 01-03-2013 |
20130011051 | CODED APERTURE IMAGING - A method of imaging encodes light from a scene by adding projective codes expressed as a product of a known projective code matrix with a known reconstruction matrix representing an image reconstruction operation. The encoded light is detected at a photodetector. The measurements are processed by compressive sensing including projective sub-sampling to represent the measurements as a linear system. The linear system is expressed as a plurality of undetermined linear equations including a product of the known reconstruction matrix and an unknown sparse vector. The sparse vector is approximated to provide solutions to the undetermined linear equations. At least one of a reconstructed image and an exploited image is generated from the measurements using solutions to the undetermined linear equations, wherein a product of the known reconstruction matrix with the solutions to underdetermined linear equations provides an image representation of the scene of interest having N pixels, where N>k. | 01-10-2013 |
20130016899 | Systems and Methods for Matching Visual Object ComponentsAANM Li; YuanAACI Los AngelesAAST CAAACO USAAGP Li; Yuan Los Angeles CA USAANM Adam; HartwigAACI Marina del ReyAAST CAAACO USAAGP Adam; Hartwig Marina del Rey CA US - Systems and methods for modeling the occurrence of common image components (e.g., sub-regions) in order to improve visual object recognition are disclosed. In one example, a query image may be matched to a training image of an object. A matched region within the training image to which the query image matches may be determined and a determination may be made whether the matched region is located within an annotated image component of the training image. When the matched region matches only to the image component, an annotation associated with the component may be identified. In another example, sub-regions within a plurality of training image corpora may be annotated as common image components including associated information (e.g., metadata). Matching sub-regions appearing in many training images of objects may be down-weighted in the matching process to reduce possible false matches to query images including common image components. | 01-17-2013 |
20130022263 | System and Method for Detecting and Tracking Features in Images - A system and method for tracking features is provided which allows for the tracking of features that move in a series of images A training set of images is processed to produce clustered shape subspaces corresponding to the set of images, such that non-linear shape manifolds in the images are represented as piecewise, overlapping linear surfaces that are clustered according to similarities in perspectives. A landmark-based training algorithm (e.g., ASM) is applied to the clustered shape subspaces to train a model of the clustered shape subspaces and to create training data. A subsequent image is processed using the training data to identify features in the target image by creating an initial shape, superimposing the initial shape on the target image, and then iteratively deforming the shape in accordance with the model until a final shape is produced corresponding to a feature in the target image. | 01-24-2013 |
20130028508 | SYSTEM AND METHOD FOR COMPUTING THE VISUAL PROFILE OF A PLACE - A system and method for computing a place profile are disclosed. The method includes providing a geographical definition of a place, retrieving a set of images based on the geographical place definition. With a classifier, image-level statistics for the retrieved images are generated. The classifier has been trained to generate image-level statistics for a finite set of classes, such as different activities. The image-level statistics are aggregated to generate a place profile for the defined place which may be displayed to a user who has provided information for generating the geographical definition or used in an application such as a recommender system or to generate a personal profile for the user. | 01-31-2013 |
20130039570 | USING EXTRACTED IMAGE TEXT - Methods, systems, and apparatus including computer program products for using extracted image text are provided. In one implementation, a computer-implemented method is provided. The method includes receiving an input of one or more image search terms and identifying keywords from the received one or more image search terms. The method also includes searching a collection of keywords including keywords extracted from image text, retrieving an image associated with extracted image text corresponding to one or more of the image search terms, and presenting the image. | 02-14-2013 |
20130039571 | METHOD AND SYSTEM FOR LEARNING A SAME-MATERIAL CONSTRAINT IN AN IMAGE - In a first exemplary embodiment of the present invention, an automated, computerized method is provided for processing an image. According to a feature of the present invention, the method comprises the steps of providing an image file depicting an image, in a computer memory, assembling a feature vector for the image file, the feature vector containing information regarding a likelihood that a selected pair of regions of the image file are of a same intrinsic characteristic, providing a classifier derived from a computer learning technique, computing a classification score for the selected pair of regions of the image file, as a function of the feature vector and the classifier and classifying the regions as being of the same intrinsic characteristic, as a function of the classification score. | 02-14-2013 |
20130039572 | METHOD AND SYSTEM FOR LOW COMPLEXITY TRANSCODING OF IMAGES WITH NEAR OPTIMAL QUALITY - Method and system for low complexity transcoding of images with near optimal quality for display on a terminal are presented. Generating effective transcoding parameters prior to transcoding and retrieving the transcoding parameters based on the features of the input image and the characteristics of the terminal, an output image quality close to that produced by optimal quality transcoding is achieved. The processing time is much smaller in comparison to that required for optimal quality transcoding. | 02-14-2013 |
20130044942 | EVENT DETECTION THROUGH PATTERN DISCOVERY - Events are classified through string pattern recognition. Text labels are assigned to image primitives in a time-ordered set of training images and to related time-ordered transactions in an associated training transaction log in a combined time-ordered training string of text labels as a function of image types. Transactions are labeled in a training transaction log with a transaction label, a training primitive image of a start of a transaction with a start image text label, a training primitive of an entry of a transaction into the log with an entry image text label, and a training primitive of a conclusion of a transaction with an ending image text label. Positive subset string patterns are discovered representing true events from the combined time-ordered training string of text labels, and negative subset string patterns defined by removing single transaction primitive labels from the positive subset string patterns. | 02-21-2013 |
20130051661 | Using Human Intelligence Tasks for Precise Image Analysis - Described are systems, methods, computer programs, and user interfaces for image location, acquisition, analysis, and data correlation that uses human-in-the-loop processing, Human Intelligence Tasks (HIT), and/or or automated image processing. Results obtained using image analysis are correlated to non-spatial information useful for commerce and trade. For example, images of regions of interest of the earth are used to count items (e.g., cars in a store parking lot to predict store revenues), detect events (e.g., unloading of a container ship, or evaluating the completion of a construction project), or quantify items (e.g., the water level in a reservoir, the area of a farming plot). | 02-28-2013 |
20130051662 | LEARNING APPARATUS, METHOD FOR CONTROLLING LEARNING APPARATUS, DETECTION APPARATUS, METHOD FOR CONTROLLING DETECTION APPARATUS AND STORAGE MEDIUM - A learning apparatus comprises a plurality of detection units configured to detect a part or whole of a target object in an image and output a plurality of detection results; an estimation unit configured to estimate a state of the target object based on at least one of the plurality of detection results; a classification unit configured to classify the image into a plurality of groups based on the state of the target object; and a weight calculation unit configured to calculate weight information on each of the plurality of detection units for each of the groups based on the detection results. | 02-28-2013 |
20130058566 | INFORMATION PROCESSOR, INFORMATION PROCESSING METHOD, AND PROGRAM - An information processor includes a detection unit detecting a photographic subject region of an image, a characteristic amount generation unit generating a characteristic amount including at least positional information of the photographic subject region for each of the detected photographic subject region, a combined characteristic amount generation unit generating a combined characteristic amount corresponding to the image by combining the characteristic amount generated for each of the photographic subject region, and an identification unit identifying a label corresponding to a combination of a photographic subject appearing in the image based on the generated combined characteristic amount. | 03-07-2013 |
20130058567 | MULTISCALE MODULUS FILTER BANK AND APPLICATIONS TO PATTERN DETECTION, CLUSTERING, CLASSIFICATION AND REGISTRATION - A digital filter bank having a number J≧1 of stages is disclosed. For each integer j such that 1≦j≦J, the j-th stage includes a plurality of filtering units ( | 03-07-2013 |
20130064444 | DOCUMENT CLASSIFICATION USING MULTIPLE VIEWS - A training system, training method, and a system and method of use of a trained classification system are provided. A classifier may be trained with a first “cheap” view but not using a second “costly” view of each of the training samples, which is not available at test time. The two views of samples are each defined in a respective original feature space. An embedding function is learned for embedding at least the first view of the training samples into a common feature space in which the second view can also be embedded or is the same as the second view original feature space. Labeled training samples (first view only) for training the classifier are embedded into the common feature space using the learned embedding function. The trained classifier can be used to predict labels for test samples for which the first view has been embedded in the common feature space with the embedding function. | 03-14-2013 |
20130077856 | PROCESSES AND SYSTEMS FOR TRAINING MACHINE TYPESETS FOR CHARACTER RECOGNITION - Processes and systems for training machine vision systems for use with OCR algorithms to recognize characters. Such a process includes identifying characters to be recognized and individually generating at least a first set of templates for each of the characters. Each template comprises a grid of cells and is generated by selecting certain cells of the grid to define a pattern that correlates to a corresponding one of the characters. Information relating to the templates is then saved on media, from which the information can be subsequently retrieved to regenerate the templates. The templates can be used in an optical character recognition algorithm to recognize at least some of the characters contained in a marking. | 03-28-2013 |
20130077857 | PRINTER IMAGE LOG SYSTEM FOR DOCUMENT GATHERING AND RETENTION - A system and method for document image acquisition and retrieval find application in litigation for responding to discovery requests. The method includes receiving automatically acquired electronic image logs comprising image data and associated records for documents processed by a plurality of image output devices within an organization. When a request for document production is received, the image logs (and/or information extracted therefrom) are automatically filtered through at least one classifier trained to return documents responsive to the document request, and documents corresponding to the filtered out image logs are output. One of the filters may be configured for filtering out documents that include attorney-client exchanges. | 03-28-2013 |
20130083996 | Using Machine Learning to Improve Visual Comparison - In some embodiments, information associated with a first plurality of image pairs is received. Each image pair is assessed to detect visual pairwise differences that qualify as an error. A visual pairwise difference may be a difference with respect to at least one of position, size, color, or style. A prediction engine is trained based upon the assessed visual pairwise differences. Information associated with a second plurality of image pairs is received. Each of these image pairs comprises at least a portion of a visual end-user experience screen of an event-driven application executed in a client-tier environment. Each of these image pairs is assessed, using the prediction engine, to detect visual pairwise differences that qualify as an error. User feedback is received, indicating that at least one assessed pairwise difference should not have qualified as an error. The prediction engine is then re-trained based on the user feedback. | 04-04-2013 |
20130108152 | IMAGE QUALITY ANALYSIS FOR SEARCHES | 05-02-2013 |
20130108153 | METHODS AND APPARATUS TO PERFORM IMAGE CLASSIFICATION BASED ON PSEUDORANDOM FEATURES | 05-02-2013 |
20130108154 | IMAGE PROCESSING LEARNING DEVICE, IMAGE PROCESSING LEARNING METHOD, AND IMAGE PROCESSING LEARNING PROGRAM | 05-02-2013 |
20130114888 | IMAGE PROCESSING APPARATUS, COMPUTER PROGRAM PRODUCT, AND IMAGE PROCESSING METHOD - According to an embodiment, an image processing apparatus includes a feature data calculator, a generating unit, and an adding unit. The feature data calculator calculates feature data representing changes in pixel values within a first range of an input image. The generating unit obtains a weight of a predetermined image pattern on the basis of a probability distribution and the feature data. The weight represents a pattern of changes in the pixel values. The probability distribution represents a distribution of relative values of feature data of a learning image containing a high-frequency component with respect to feature data of a learning image. The generating unit weights the predetermined image pattern with the weight so as to generate a high-frequency component with respect to the input image. The adding unit adds the high-frequency component to the input image. | 05-09-2013 |
20130121565 | Method and Apparatus for Local Region Selection - Methods and apparatus for local region selection are described. A scribble-based, edge-aware local region selection tool or module that implements a local region selection method may allow a user to draw scribbles or strokes indicating different classes of content. The method may train Gaussian mixture models (GMMs) for each class from the user input. The GMMs may be applied to the image to generate a probability map for each class. Post-processing may be optionally performed to remove structural outliers. The probability maps may be smoothed using a geodesic smoothing technique. A geodesic smoothing technique may be applied that considers other classes when smoothing each class to limit or prevent propagation of a region corresponding to the class into other regions corresponding to other classes. The smoothed probability maps may be combined to generate a final region selection mask. | 05-16-2013 |
20130121566 | Automatic Image Adjustment Parameter Correction - Techniques are disclosed relating to modifying an automatically predicted adjustment. In one embodiment, the automatically predicted adjustment may be adjusted, for example, based on a rule. The automatically predicted adjustment may be based on a machine learning prediction. A new image may be globally adjusted based on the modified automatically predicted adjustment. | 05-16-2013 |
20130129198 | SMART 3D PACS WORKFLOW BY LEARNING - Methods and systems to provide a hanging protocol including three-dimensional manipulation for display of clinical images in an exam are disclosed. An example method includes detecting selection of a new image exam for display by a user. The example method includes automatically identifying at least one of a) a previously learned hanging protocol saved for the user and b) a saved hanging protocol associated with a prior image exam corresponding to the new image exam. The example method includes applying the saved hanging protocol to the new image exam, the saved hanging protocol including three-dimensional manipulation to be automatically applied to the new image exam as part of the hanging protocol configuration for display. The example method includes facilitating display of the new image exam based on the saved hanging protocol. | 05-23-2013 |
20130129199 | OBJECT-CENTRIC SPATIAL POOLING FOR IMAGE CLASSIFICATION - A method is provided for classifying an image. The method includes inferring location information of an object of interest in an input representation of the image. The method further includes determining foreground object features and background object features from the input representation of the image. The method additionally includes pooling the foreground object features separately from the background object features using the location information to form a new representation of the image. The new representation is different than the input representation of the image. The method also includes classifying the image based on the new representation of the image. | 05-23-2013 |
20130129200 | DEVICE FOR SETTING IMAGE ACQUISITOIN CONDITIONS, AND COMPUTER PROGRAM - The present invention relates to a device ( | 05-23-2013 |
20130142417 | SYSTEM AND METHOD FOR AUTOMATICALLY DEFINING AND IDENTIFYING A GESTURE - A system and method for creating a gesture and generating a classifier that can identify the gesture for use with an application is described. The designer constructs a training set of data containing positive and negative examples of the gesture. Machine learning algorithms are used to compute the optimal classification of the training data into positive and negative instances of the gesture. The machine learning algorithms generate a classifier which, given input data, makes a decision on whether the gesture was performed in the input data or not. | 06-06-2013 |
20130142418 | RANKING AND SELECTING REPRESENTATIVE VIDEO IMAGES - Techniques are described herein for selecting representative images for video items using a trained machine learning engine. A training set is fed to a machine learning engine. The training set includes, for each image in the training set, input parameter values and an externally-generated score. Once a machine learning model has been generated based on the training set, input parameters for unscored images are fed to the trained machine learning engine. Based on the machine learning model, the trained machine learning engine generates scores for the images. To select a representative image for a particular video item, candidate images for that particular video item may be ranked based on their scores, and the candidate image with the top score may be selected as the representative image for the video item. | 06-06-2013 |
20130142419 | ENCODER OPTIMIZATION OF ADAPTIVE LOOP FILTERS IN HEVC - An optimized adaptive loop filter does not redesign filters inside the optimization loop of signaling depth which saves computations. Additionally, the Sum of Squared Errors (SSE) (distortion) of blocks is computed for the smallest blocks, thus, allowing for the distortion of larger blocks to be computed efficiently by adding block SSEs together which saves computations by removing redundant operations to calculate SSE of a block each time. | 06-06-2013 |
20130142420 | IMAGE RECOGNITION INFORMATION ATTACHING APPARATUS, IMAGE RECOGNITION INFORMATION ATTACHING METHOD, AND NON-TRANSITORY COMPUTER READABLE MEDIUM - An image recognition information attaching apparatus includes a retrieving unit that retrieves image information on a per piece basis of identification information, from the image information having the identification information associated thereto in advance, a generator unit that generates feature information from the image information retrieved by the retrieving unit, and a learning unit that provides a learning result by learning a relation between the feature information generated by the generator unit and the identification information of the image information corresponding to the feature information, using a stochastic model including a mixture of a plurality of probability distributions. | 06-06-2013 |
20130142421 | Method for Fast, Robust, Multi-Dimensional Pattern Recognition - A method and system for probe-based pattern matching including an apparatus for synthetic training of a model of a pattern. The apparatus comprises a sensor for obtaining an image of the pattern and a processor for receiving the image of the pattern from the sensor and running a program. In the steps performed by the program a boundary of the pattern in the image is identified. A plurality of positive probes are placed at selected points along the boundary of the pattern and at least one straight segment of the boundary of the pattern is identified. The at least one straight segment of the boundary is extended to provide an imaginary straight segment and a plurality of negative probes are placed at selected points along the imaginary straight segment, where each negative probe has a negative weight. | 06-06-2013 |
20130142422 | IMAGE PROCESSING METHOD, AND IMAGE PROCESSOR - Included are (a) performing processes on second training data items stored in a training database to generate third training data items each obtained through a corresponding one of the processes, (b) selecting, from among the third training data items generated in step (a), a selection data item having a highest similarity to a feature data item of the input image, (c) generating a high-frequency data item by: determining (i) the second training data item used in generating the selection data item and (ii) a first process performed on the second training data item to generate the selection data item; and performing the first process on the first training data item that is paired with the determined second training data item; and (d) generating an output image by adding an image indicated by the high-frequency data item to the input image. | 06-06-2013 |
20130142423 | IMAGE CLUSTERING USING A PERSONAL CLOTHING MODEL - The disclosure relates to a system and a method for generating clothing feature data representative of at least one clothing feature of a piece of clothing being worn by the person in a set of images, and training a discriminative clothing classifier using the clothing feature data to provide a personal clothing model that corresponds to the piece of clothing. The personal clothing model can be used to identify additional images in which the person appears. | 06-06-2013 |
20130142424 | Optical Pattern Recognition Technique - Disclosed is a distortion invariant system, method and computer readable medium for detecting the presence of one or more predefined targets in an input image. The input image and a synthetic discriminant function (SDF) reference image are correlated in a shift phase-encoded fringe-adjusted joint transform correlation (SPFJTC) correlator yielding a correlation output. A peak-to-clutter ratio (PCR) is determined for the correlation output and compared to a threshold value. A predefined target is present in the input image when the PCR is greater than or equal to the threshold value. | 06-06-2013 |
20130148880 | Image Cropping Using Supervised Learning - Software for supervised learning extracts a set of pixel-level features from each source image in collection of source images. Each of the source images is associated with a thumbnail created by an editor. The software also generates a collection of unique bounding boxes for each source image. And the software calculates a set of region-level features for each bounding box. Each region-level feature results from the aggregation of pixel values for one of the pixel-level features. The software learns a regression model, using the calculated region-level features and the thumbnail associated with the source image. Then the software chooses a thumbnail from a collection of unique bounding boxes in a new image, based on application of the regression model. | 06-13-2013 |
20130148881 | Image Classification - The present disclosure introduces a method and an apparatus for classifying images. Classification image features of an image for classification are extracted. Based on a similarity relationship between each classification image feature and one or more visual words in a pre-generated visual dictionary, each classification image feature is quantified by multiple visual words in the visual dictionary and a similarity coefficient between each classification image feature and each of the visual words is determined. Based on the similarity coefficient of each visual word that corresponds to different classification image features, a weight of each visual word is determined to establish a classification visual word histogram of the image for classification. The classification visual word histogram is input into an image classifier that is trained by sample visual word histograms arising from multiple sample images. An output result is used to determine a classification of the image for classification. | 06-13-2013 |
20130156297 | Learning Image Processing Tasks from Scene Reconstructions - Learning image processing tasks from scene reconstructions is described where the tasks may include but are not limited to: image de-noising, image in-painting, optical flow detection, interest point detection. In various embodiments training data is generated from a 2 or higher dimensional reconstruction of a scene and from empirical images of the same scene. In an example a machine learning system learns at least one parameter of a function for performing the image processing task by using the training data. In an example, the machine learning system comprises a random decision forest. In an example, the scene reconstruction is obtained by moving an image capture apparatus in an environment where the image capture apparatus has an associated dense reconstruction and camera tracking system. | 06-20-2013 |
20130156298 | Using High-Level Attributes to Guide Image Processing - Using high-level attributes to guide image processing is described. In an embodiment high-level attributes of images of people such as height, torso orientation, body shape, gender are used to guide processing of the images for various tasks including but not limited to joint position detection, body part classification, medical image analysis and others. In various embodiments one or more random decision forests are trained using images where global variable values such as player height are known in addition to ground-truth data appropriate for the image processing task concerned. In some examples sequences of images are used where global variables are static or vary smoothly over the sequence. In some examples one or more trained random decision forests are used to find global variable values as well as output values for the task concerned such as joint positions or body part classes. | 06-20-2013 |
20130156299 | METHOD AND APPARATUS FOR DETECTING PEOPLE WITHIN VIDEO FRAMES BASED UPON MULTIPLE COLORS WITHIN THEIR CLOTHING - A video analytic device performs a method for detecting people within frames of video based upon multiple colors within their clothing. The method includes: receiving a frame of video; and determining that a first color region within the frame matches a first color of interest for a clothing uniform, wherein the determining is based on a first set of color representation constraints. The method further includes determining that a second color region within the frame matches a second color of interest for the clothing uniform, wherein the determining is based on a second set of color representation constraints and the first and second colors of interest are different. In addition, the method includes applying a set of geometric constraints to the first and second color regions to determine a count of people within the frame wearing the clothing uniform. | 06-20-2013 |
20130156300 | Multi-Class Classification Method - A test sample is classified by determining a nearest subspace residual from subspaces learned from multiple different classes of training samples, and a collaborative residual from a collaborative representation of a dictionary constructed from all of the test samples. The residuals are used to determine a regularized residual. The subspaces, the dictionary and the regularized residual are inputted into a classifier, wherein the classifier includes a collaborative representation classifier and a nearest subspace classifier, and a label is assigned to the test sample using the classifier, and wherein the regularization parameter balances a trade-off between the collaborative representation classifier the nearest subspace classifier. | 06-20-2013 |
20130156301 | METHOD AND SYSTEM FOR RECOGNIZING IMAGES - A method and a system for recognizing at least one testing image according to classes are provided, wherein each of the classes includes sample images. The method includes generating an average image of each class according to the sample images, generating a feature enhancement mask according to differences between the average images of the classes, enhancing the sample images of each class by using the feature enhancement mask to generate a plurality of enhanced sample images corresponding to each class, and training a classifier according to the enhanced sample images of each class. The method also includes enhancing the at least one testing image by using the feature enhancement mask to generate an enhanced testing image, classifying the enhanced testing image into one of the classes by using the classifier, and recognizing that the testing image belongs to the classified class. Thereby, this method can effectively recognize the testing image. | 06-20-2013 |
20130156302 | HANDWRITTEN WORD SPOTTER SYSTEM USING SYNTHESIZED TYPED QUERIES - A wordspotting system and method are disclosed for processing candidate word images extracted from handwritten documents. In response to a user inputting a selected query string, such as a word to be searched in one or more of the handwritten documents, the system automatically generates at least one computer-generated image based on the query string in a selected font or fonts. A model is trained on the computer-generated image(s) and is thereafter used in the scoring the candidate handwritten word images. The candidate or candidates with the highest scores and/or documents containing them can be presented to the user, tagged, or otherwise processed differently from other candidate word images/documents. | 06-20-2013 |
20130156303 | IMAGE PROCESSING APPARATUS AND IMAGE PROCESSING METHOD - An image processing method is provided for an image processing apparatus which executes processing by allocating a plurality of weak discriminators to form a tree structure having branches corresponding to types of objects so as to detect objects included in image data. Each weak discriminator calculates a feature amount to be used in a calculation of an evaluation value of the image data, and discriminates whether or not the object is included in the image data by using the evaluation value. The weak discriminator allocated to a branch point in the tree structure further selects a branch destination using at least some of the feature amounts calculated by weak discriminators included in each branch destination. | 06-20-2013 |
20130156304 | METHOD FOR CLASSIFICATION OF VIDEOS - A method for classifying a video regarding a subjective characteristic, the method comprising:
| 06-20-2013 |
20130163859 | REGRESSION TREE FIELDS - A new tractable model solves labeling problems using regression tree fields, which represent non-parametric Gaussian conditional random fields. Regression tree fields are parameterized by non-parametric regression trees, allowing universal specification of interactions between image observations and variables. The new model uses regression trees corresponding to various factors to map dataset content (e.g., image content) to a set of parameters used to define the potential functions in the model. Some factors define relationships among multiple variable nodes. Further, the training of regression trees is scalable, both in the training set size and in the fact that the training can be parallelized. In one implementation, maximum pseudolikelihood learning provides for joint training of various aspects of the model, including feature test selection and ordering (i.e., the structure of the regression trees), parameters of each factor in the graph, and the scope of the interacting variable nodes used in the graph. | 06-27-2013 |
20130163860 | Information Processing Device, Information Processing Method and Program - The present invention relates to an information processing device, an information processing method, and a program capable of easily adding an annotation to content. | 06-27-2013 |
20130170738 | COMPUTER-IMPLEMENTED METHOD, A COMPUTER PROGRAM PRODUCT AND A COMPUTER SYSTEM FOR IMAGE PROCESSING - The present description refers in particular to a computer-implemented method, a computer program product and a computer system for image processing, the method comprising: receiving at least one user image; identifying a plurality of image classification elements of the user image by: assigning an initial classification to the user image, wherein the initial classification is based on temporal data associated with the user image; determining at least one image label that globally describes content of the user image; calculating a label correctness value for each image label; recognizing at least one image component of the user image; calculating a component correctness value for each image component; correlating the image label and the image component using the label correctness value and the component correctness value, whereby a correlated image label and a correlated image component are identified; applying a rule to determine a category of the user image, wherein the rule is based on at least one of the following: the temporal data, the correlated image label and the correlated image component; and producing a final classification of the user image including the following image classification elements: the initial classification, the correlated image label, the correlated image component, and the category. | 07-04-2013 |
20130170739 | LEARNING APPARATUS, A LEARNING SYSTEM, LEARNING METHOD AND A LEARNING PROGRAM FOR OBJECT DISCRIMINATION - A learning apparatus in the present invention includes a weak discriminator generation unit that generates a weak discriminator which calculates a discrimination score of an instance of a target based on a feature and a bag label, a weak discrimination unit which calculates the discrimination score based on the generated weak discriminator, an instance probability calculation unit that calculates an instance probability of the target instance based on the calculated the discrimination score, a bag probability calculation unit that calculates a probability that no smaller than two positive instances are included in the bag based on the calculated instance probability and a likelihood calculation unit which calculates likelihood representing plausibility of the bag probability based on the bag label. | 07-04-2013 |
20130182946 | METHODS AND SYSTEM FOR ANALYZING AND RATING IMAGES FOR PERSONALIZATION - As set forth herein, a computer-implemented method facilitates pre-analyzing an image and automatically suggesting to the user the most suitable regions within an image for text-based personalization. Image regions that are spatially smooth and regions with existing text (e.g. signage, banners, etc.) are primary candidates for personalization. This gives rise to two sets of corresponding algorithms: one for identifying smooth areas, and one for locating text regions. Smooth regions are found by dividing the image into blocks and applying an iterative combining strategy, and those regions satisfying certain spatial properties (e.g. size, position, shape of the boundary) are retained as promising candidates. In one embodiment, connected component analysis is performed on the image for locating text regions. Finally, based on the smooth and text regions found in the image, several alternative approaches are described herein to derive an overall metric for “suitability for personalization.” | 07-18-2013 |
20130182947 | APPARATUS AND METHOD FOR ESTIMATING POSE OF OBJECT - An apparatus and method for estimating a pose of an object are provided. The apparatus includes an object input unit configured to input an object in an object tracking unit and an object identifying unit, an object tracking unit configured to obtain a tracked pose probability density of the object based on a tracking scheme, an object identifying unit configured to obtain an identified pose probability density of the object based on a training model, and a combination unit configured to obtain an estimated pose probability density of the object using a combination of the tracked pose probability density and the identified pose probability density and to estimate a pose of the object based on the estimated pose probability density of the object. Through the combination, a cumulative error occurring in the object tracking may be corrected, resulting in more accurate object estimation. | 07-18-2013 |
20130182948 | Method and Apparatus for Training a Probe Model Based Machine Vision System - A method for training a pattern recognition algorithm including the steps of identifying the known location of the pattern that includes repeating elements within a fine resolution image, using the fine resolution image to train a model associated with the fine image, using the model to examine the fine image resolution image to generate a score space, examining the score space to identify a repeating pattern frequency, using a coarse image that is coarser than the finest image resolution image to train a model associated with the coarse image, using the model associated with the coarse image to examine the coarse image thereby generating a location error, comparing the location error to the repeating pattern frequency and determining if the coarse image resolution is suitable for locating the pattern within a fraction of one pitch of the repeating elements. | 07-18-2013 |
20130195351 | IMAGE PROCESSOR, IMAGE PROCESSING METHOD, LEARNING DEVICE, LEARNING METHOD AND PROGRAM - Disclosed herein is an image processor including: a feature point extraction section adapted to extract the feature points of an input image; a correspondence determination section adapted to determine the correspondence between the feature points of the input image and those of a reference image using a feature point dictionary; a feature point coordinate distortion correction section adapted to correct the coordinates of the feature points of the input image corresponding to those of the reference image; a projection relationship calculation section adapted to calculate the projection relationship between the input and reference images; a composite image coordinate transform section adapted to generate a composite Image to be attached from a composite image; and an output image generation section adapted to merge the input image with the composite image to be attached. | 08-01-2013 |
20130202198 | Landmarks from Digital Photo Collections - Methods and systems for automatic detection of landmarks in digital images and annotation of those images are disclosed. A method for detecting and annotating landmarks in digital images includes the steps of automatically assigning a tag descriptive of a landmark to one or more images in a plurality of text-associated digital images to generate a set of landmark-tagged images, learning an appearance model for the landmark from the set of landmark-tagged images, and detecting the landmark in a new digital image using the appearance model. The method can also include a step of annotating the new image with the tag descriptive of the landmark. | 08-08-2013 |
20130202199 | USING HIGHER ORDER STATISTICS TO ESTIMATE PIXEL VALUES IN DIGITAL IMAGE PROCESSING TO IMPROVE ACCURACY AND COMPUTATION EFFICIENCY - A method, system and computer program product for improving accuracy and computation efficiency in interpolation, upsampling and color channel estimation. A Bayesian estimator used to estimate the value of a pixel in an image is constructed using measurements of high-order (e.g., 3rd, 4th, 5th) statics for nearby points in natural images. These measurements reveal highly systematic statistical regularities that were ignored from the prior algorithms due to their restrictive measurements and assumptions. As a result, the accuracy in interpolation, upsampling and color channel prediction is improved. Furthermore, the process for constructing a Bayesian estimator is simpler and more direct by storing in a table the mean value of the pixel value to be estimated for each combination of values of nearby points in training samples. As a result of having a simpler and more direct approach than existing methods, the computational efficiency is improved. | 08-08-2013 |
20130208977 | RECEPTIVE FIELD LEARNING FOR POOLED IMAGE FEATURES - Systems and methods are disclosed for image classification by receiving an overcomplete set of spatial regions, jointly optimizing the classifier and the pooling region for each pooled feature; and performing incremental feature selection and retraining using a grafting process to efficiently train the classifier. | 08-15-2013 |
20130208978 | CONTINUOUS CHARTING OF NON-UNIFORMITY SEVERITY FOR DETECTING VARIABILITY IN WEB-BASED MATERIALS - A computerized inspection system is described for detecting the presence of non-uniformity defects in a manufactured web material and for providing output indicative of a severity level of each defect. The system provides output that provides the severity levels of the non-uniformity defects in real-time on a continuous scale. Training software processes a plurality of training samples to generate a model, where each of the training samples need only be assigned one of a set of discrete rating labels for the non-uniformity defects. The training software generates the model to represent a continuous ranking of the training images, and the inspection system utilizes the model to compute the severity levels of the web material on a continuous scale in real-time without limiting the output to the discrete rating labels assigned to the training samples. | 08-15-2013 |
20130216127 | IMAGE SEGMENTATION USING REDUCED FOREGROUND TRAINING DATA - Methods of image segmentation using reduced foreground training data are described. In an embodiment, the foreground and background training data for use in segmentation of an image is determined by optimization of a modified energy function. The modified energy function is the energy function used in image segmentation with an additional term comprising a scalar value. The optimization is performed for different values of the scalar to produce multiple initial segmentations and one of these segmentations is selected based on pre-defined criteria. The training data is then used in segmenting the image. In other embodiments further methods are described: one places an ellipse inside the user-defined bounding box to define the background training data and another uses a comparison of properties of neighboring image elements, where one is outside the user-defined bounding box, to reduce the foreground training data. | 08-22-2013 |
20130223726 | METHOD AND APPARATUS OF CLASSIFICATION AND OBJECT DETECTION, IMAGE PICKUP AND PROCESSING DEVICE - A maximum hypersphere is created in the feature space according to support vectors, wherein the support vectors are one or more feature vectors in a feature space. A center of the created maximum hypersphere is calculated according to the support vector(s). A decision hyper sphere is created with the same center as the calculated center of the created maximum hypersphere. Feature vector(s) are classified within the decision hypersphere, as positive feature vector(s). False positive rate is kept to a predetermined level to provide effective object detection. | 08-29-2013 |
20130223727 | METHOD AND DEVICE FOR LEARNING OF A CLASSIFIER, AND PROCESSING APPARATUS - Unlabeled samples are added to existing samples. Candidate samples for new support vectors are extracted from the added unlabeled samples using a plurality of existing support vectors. The new support vectors are selected from the candidate samples using the plurality of existing support vectors. | 08-29-2013 |
20130230236 | IMAGE RETRIEVAL APPARATUS - Embodiments describe an image retrieval apparatus. The image retrieval apparatus includes an unlabelled image selector for selecting one or more unlabelled image(s) from an image database; and a main learner for training in each feedback round of the image retrieval, estimating relevance of images in the image database and a user's intention, and determining retrieval results, wherein the main learner makes use of the unlabelled image(s) selected by the unlabelled image selector in the estimation. In addition, the image retrieval apparatus may also include an active selector for selecting, in each feedback round and according to estimation results of the main learner, one or more unlabelled image(s) from the image database for the user to label. | 09-05-2013 |
20130243308 | INTEGRATED INTERACTIVE SEGMENTATION WITH SPATIAL CONSTRAINT FOR DIGITAL IMAGE ANALYSIS - An integrated interactive segmentation with spatial constraint method utilizes a combination of several of the most popular online learning algorithms into one and implements a spatial constraint which defines a valid mask local to the user's given marks. Additionally, both supervised learning and statistical analysis are integrated, which are able to compensate each other. Once prediction and activation are obtained, pixel-wised multiplication is conducted to fully indicate how likely each pixel belongs to the foreground or background. | 09-19-2013 |
20130243309 | SYSTEM AND METHOD FOR AUTOMATIC LANDMARK LABELING WITH MINIMAL SUPERVISION - A system and method for estimating a set of landmarks for a large image ensemble employs only a small number of manually labeled images from the ensemble and avoids labor-intensive and error-prone object detection, tracking and alignment learning task limitations associated with manual image labeling techniques. A semi-supervised least squares congealing approach is employed to minimize an objective function defined on both labeled and unlabeled images. A shape model is learned on-line to constrain the landmark configuration. A partitioning strategy allows coarse-to-fine landmark estimation. | 09-19-2013 |
20130251244 | REAL TIME HEAD POSE ESTIMATION - Methods are provided for generating a low dimension pose space and using the pose space to estimate one or more head rotation angles of a user head. In one example, training image frames including a test subject head are captured under a plurality of conditions. For each frame an actual head rotation angle about a rotation axis is recorded. In each frame a face image is detected and converted to an LBP feature vector. Using principal component analysis a PCA feature vector is generated. Pose classes related to rotation angles about a rotation axis are defined. The PCA feature vectors are grouped into a pose class that corresponds to the actual rotation angle associated with the PCA feature vector. Linear discriminant analysis is applied to the pose classes to generate the low dimension pose space. | 09-26-2013 |
20130251245 | Method for Reducing Blocking Artifacts in Images - Blocking artifacts are reduced by projecting each patch obtained from an input image onto a set of bases vectors to determine multiple representations for each patch. The set of bases vectors are learned from a training image, and the bases vectors include a full basis vector, and one or two subspace bases vectors. An optimal basis vector is determined in the set of bases vectors for each patch according to the projection. A threshold is applied to coefficients of the optimal basis vector to determine a filtered representation for each patch, and a reconstructed patch is generated using the filtered representation. Then, the aggregating the reconstructed patches are aggregated to produce an output image. | 09-26-2013 |
20130251246 | METHOD AND A DEVICE FOR TRAINING A POSE CLASSIFIER AND AN OBJECT CLASSIFIER, A METHOD AND A DEVICE FOR OBJECT DETECTION - A method and a device for training a pose classifier and an object classifier, and a method and a device for objection detection, relating to the field of image processing are provided. The object detection method includes acquiring input image samples; performing pose estimation processing on said input image sample according to said pose classifier; and performing object detection on the processed input image sample according to said pose classifier to acquire the location information of the object, wherein said object is an object with joints. Objects in different poses can be detected and therefore the object hit rate is increased. | 09-26-2013 |
20130251247 | SYSTEM AND METHODS FOR ARABIC TEXT RECOGNITION AND ARABIC CORPUS BUILDING - A method for automatically recognizing Arabic text includes building an Arabic corpus comprising Arabic text files written in different writing styles and ground truths corresponding to each of the Arabic text files, storing writing-style indices in association with the Arabic text files, digitizing a line of Arabic characters to form an array of pixels, dividing the line of the Arabic characters into line images, forming a text feature vector from the line images, training a Hidden Markov Model using the Arabic text files and ground truths in the Arabic corpus in accordance with the writing-style indices, and feeding the text feature vector into a Hidden Markov Model to recognize the line of Arabic characters. | 09-26-2013 |
20130251248 | ENHANCED MAX MARGIN LEARNING ON MULTIMODAL DATA MINING IN A MULTIMEDIA DATABASE - Multimodal data mining in a multimedia database is addressed as a structured prediction problem, wherein mapping from input to the structured and interdependent output variables is learned. A system and method for multimodal data mining is provided, comprising defining a multimodal data set comprising image information; representing image information of a data object as a set of feature vectors in a feature space; clustering in the feature space to group similar features; associating a non-image representation with a respective image data object based on the clustering; determining a joint feature representation of a respective data object as a mathematical weighted combination of a set of components of the joint feature representation; optimizing a weighting for a plurality of components of the mathematical weighted combination with respect to a prediction error between a predicted classification and a training classification; and employing the mathematical weighted combination for automatically classifying a new data object. | 09-26-2013 |
20130279800 | LEARNING PART-BASED MODELS OF OBJECTS - A system and method are disclosed for learning part-based object models during a learning phase from training images and applying the learned object models to an input image during runtime. The learned part-based object models are augmented by appearance-based models of the objects. The part-based object models correspond to the shapes of the parts of an object. The appearance-based models provide additional appearance cues to the object models for object classification. The approach to learning part-based object models has the capability of learning object models without using viewpoint labels of the objects. The learning is also invariant to scale and in-plane rotation of the objects. | 10-24-2013 |
20130294685 | MATERIAL RECOGNITION FROM AN IMAGE - A method of operating a computer system to perform material recognition based on multiple features extracted from an image is described. A combination of low-level features extracted directly from the image and multiple novel mid-level features extracted from transformed versions of the image are selected and used to assign a material category to a single image. The novel mid-level features include non-reflectance based features such as the micro-texture features micro-jet and micro-SIFT and the shape feature curvature, and reflectance-based features including edge slice and edge ribbon. An augmented Latent Dirichlet Allocation (LDA) model is provided as an exemplary Bayesian framework for selecting a subset of features useful for material recognition of objects in an image. | 11-07-2013 |
20130294686 | OPTIMAL GRADIENT PURSUIT FOR IMAGE ALIGNMENT - A method for image alignment is disclosed. In one embodiment, the method includes acquiring a facial image of a person and using a discriminative face alignment model to fit a generic facial mesh to the facial image to facilitate locating of facial features. The discriminative face alignment model may include a generative shape model component and a discriminative appearance model component. Further, the discriminative appearance model component may have been trained to estimate a score function that minimizes the angle between a gradient direction and a vector pointing toward a ground-truth shape parameter. Additional methods, systems, and articles of manufacture are also disclosed. | 11-07-2013 |
20130301910 | EXTRACTING OBJECT EDGES FROM IMAGES - A computer system may elicit from a human observer ground truth data useful in automatically detecting one or more features in images. The elicitation may include presenting an image to a human observer that has a visual indicator in an image, the visual indicator indicating having a location and orientation with respect to the image; asking the human observer to judge whether a particular image feature is present in the image at the location and orientation indicated by the visual indicator; receiving input from the human observer indicative of whether the particular image feature is present at the location and orientation indicated by the visual indicator; storing the input received from the human observer as part of the human-labeled ground truth data; and repeating the process described above one or more times in connection with a visual indicator that has a different location or orientation with respect to the image or that uses a different image. The stored human-labeled ground truth data may have a content that is useful in automatically detecting one or more features in other images. | 11-14-2013 |
20130308852 | SYSTEM AND METHOD FOR ROBUST ESTIMATION OF COLOR DEPENDENT MEASUREMENTS - Methods, devices, and computer program products for robust estimation of color-dependent measurements are described herein. In one aspect, a method for generating a reference color grid that may be placed beside a color-dependent measuring device is disclosed. The reference color grid may contain a number of colors which enable a mapping from the color space of a testing device to a reference color space. This mapping may allow a function that is able to determine an estimate of a color-dependent measurement based on a color in the reference color space to be used. In another aspect, a method for robust estimation of color-dependent measurement using a reference color guide is disclosed. | 11-21-2013 |
20130308853 | SYSTEM AND METHOD FOR SYNTHESIZING PORTRAIT SKETCH FROM A PHOTO - The present invention discloses a system and method for synthesizing a portrait sketch from a photo. The method includes: dividing the photo into a set of photo patches; determining first matching information between each of the photo patches and training photo patches pre-divided from a set of training photos; determining second matching information between each of the photo patches and training sketch patches pre-divided from a set of training sketches; determining a shape prior for the portrait sketch to be synthesized; determining a set of matched training sketch patches for each of the photo patches based on the first and the second matching information and the shape prior; and synthesizing the portrait sketch from the determined matched training sketch patches. | 11-21-2013 |
20130308854 | IMAGE PROCESSING DEVICE AND METHOD, LEARNING DEVICE AND METHOD, AND PROGRAM - There is provided an image processing device including a weight calculation unit that calculates a weight corresponding to each of a plurality of pixel values centering on a pixel of interest of an input image based on a feature amount calculated based on the plurality of pixel values centering on the pixel of interest, a regression coefficient reading unit that reads a regression coefficient stored for each class code determined based on a plurality of pixel values corresponding to the pixel of interest of the input image, and a pixel value calculation unit that calculates a pixel value of a pixel of interest of an output image by performing calculation using the plurality of pixel values, the weights, and the regression coefficients centering on the pixel of interest of the input image. | 11-21-2013 |
20130308855 | Smile Detection Techniques - Techniques are disclosed that involve the detection of smiles from images. Such techniques may employ local-binary pattern (LBP) features and/or multi-layer perceptrons (MLP) based classifiers. Such techniques can be extensively used on various devices, including (but not limited to) camera phones, digital cameras, gaming devices, personal computing platforms, and other embedded camera devices. | 11-21-2013 |
20130315477 | IMAGE SELECTION BASED ON PHOTOGRAPHIC STYLE - A system and method are disclosed for image selection based on photographic style in which photographic style annotations are learned using a data-driven approach. The method includes assigning a style value for each of a set of photographic style categories to each of a set of database images with a trained classifier of a computing device. A user's selection of a subset of the photographic style categories, such as three style categories, is received. A user interface is generated for assigning values to each of the selected photographic style categories. A set of database images is identified, based on the assigned values for each of the selected photographic style categories and the style values for each of the selected photographic style categories of the database images. | 11-28-2013 |
20130315478 | Classifying Blur State of Digital Image Pixels - A blur classification module may compute the probability that a given pixel in a digital image was blurred using a given two-dimensional blur kernel, and may store the computed probability in a blur classification probability matrix that stores probability values for all combinations of image pixels and the blur kernels in a set of likely blur kernels. Computing these probabilities may include computing a frequency power spectrum for windows into the digital image and/or for the likely blur kernels. The blur classification module may generate a coherent mapping between pixels of the digital image and respective blur states, and/or may perform a segmentation of the image into blurry and sharp regions, dependent on values stored in the matrix. Input image data may be pre-processed. Blur classification results may be employed in image editing operations to automatically target image subjects or background regions, or to estimate the depth of image elements. | 11-28-2013 |
20130315479 | Automatic Adaptation to Image Processing Pipeline - Techniques are disclosed relating to generating generic labels, translating generic labels to image pipeline-specific labels, and automatically adjusting images. In one embodiment, generic labels may be generated. Generic algorithm parameters may be generated based on training a regression algorithm with the generic labels. The generic labels may be translated to pipeline-specific labels, which may be usable to automatically adjust an image. | 11-28-2013 |
20130322740 | Method of Automatically Training a Classifier Hierarchy by Dynamic Grouping the Training Samples - The present invention uses dynamic grouping to divide up training samples to train different classification nodes. At the beginning of the training, all samples are in the same group. A clustering process is applied in the feature space of the selected feature vectors with cluster indexes accumulated. The average of all the accumulated cluster indexes is used as the threshold for splitting the samples into two groups. When the splitting criterion is met, samples are split into two groups based on their similarity in the feature space. | 12-05-2013 |
20130322741 | Teachable pattern scoring method - A computerized teachable pattern scoring method receives a teaching image and region pattern labels. A region segmentation is performed using the teaching image to generate regions of interest output. A feature measurement is performed using the teaching image and the regions of interest to generate region features output. A pattern score learning is performed using the region features and the region pattern labels to generate pattern score recipe output. A computerized region classification method using the region features and the pattern score recipe to generate pattern scores output. A region classification is performed using the pattern scores and region features to generate region class output. | 12-05-2013 |
20130322742 | Tactical Object Finder - A detection system includes processing circuitry configured to receive overhead image data divided into a plurality of image chips and receive metadata associated with the image data. The metadata includes ground sample distance information associated with the image data and provides an indication of ground area represented by each pixel within the image chips. The processing circuitry is further configured to screen the image chips for candidate detections based on a multi-stage screening process and determine whether to classify candidate detections as target detections. The process includes an intensity based screening stage, an object extraction stage that employs binary shape features to extract objects from detect positions identified based on an output of the intensity based screening stage, and a candidate detection identification stage employing template based and structural feature criteria to identify candidate detections from an output of the object extraction stage. | 12-05-2013 |
20130322743 | MULTI-CLASS IDENTIFIER, METHOD, AND COMPUTER-READABLE RECORDING MEDIUM - A multi-class identifier identifies a kind of an imager, and identifies in detail with respect to a specified kind of a group. The multi-class identifier includes: an identification fault counter providing the image for test that includes any of class labels to the kind identifiers so that the kind identifiers individually identify the kind of the provided image, and counting, for a combination of arbitrary number of kinds among the plurality of kinds, the number of times of incorrect determination in the arbitrary number of kinds that belongs to the combination; a grouping processor, for a group of the combination for which count result is equal to or greater than a predetermined threshold, adding a group label corresponding to the group to the image for learning that includes the class label corresponding to any of the arbitrary number of kinds that belongs to the group. | 12-05-2013 |
20130322744 | Method of Detecting and Identifying Substances or Mixtures and Determining Their Characteristics - This invention relates to a method of non-contact detection and identification of the type of different substances and mixtures as well as determining their characteristics as concentration, hardness, etc. The method comprises irradiation of the inspected object by a wave pulse or a series of such pulses; reception ( | 12-05-2013 |
20130329986 | SYSTEM AND METHOD FOR OPTIMIZING THE NUMBER OF CONDITIONING DATA IN MULTIPLE POINT STATISTICS SIMULATION - A computer system and a computer-implemented method for optimizing the number of conditioning data used in multiple point statistics simulation. The method includes inputting a training image representative of subsurface geological heterogeneity; and inputting an initial conservative number of conditioning data. The method further includes selecting a geometry of a template wherein a size of the template is defined by the conservative number of conditioning data; building a search tree using the template by scanning the training image with the template and storing data patterns present in the training image in the search tree to obtain a plurality of patterns; and determining a threshold number of conditioning data smaller than the initial conservative number of conditioning data beyond which estimated facies probabilities are not significantly modified by additional number of conditioning data. | 12-12-2013 |
20130329987 | VIDEO SEGMENTATION METHOD - A system and method implemented as a software tool for foreground segmentation of video sequences in real-time, which uses two Competing 1-class Support Vector Machines (C-1SVMs) operating to separately identify background and foreground. A globalized, weighted optimizer may resolve unknown or boundary conditions following convergence of the C-1SVMs. The objective of foreground segmentation is to extract the desired foreground object from live input videos, with fuzzy boundaries captured by freely moving cameras. The present disclosure proposes the method of training and maintaining two competing classifiers, based on Competing 1-class Support Vector Machines (C-1SVMs), at each pixel location, which model local color distributions for foreground and background, respectively. By introducing novel acceleration techniques and exploiting the parallel structure of the algorithm (including reweighing and max-pooling of data), real-time processing speed is achieved for VGA-sized videos. | 12-12-2013 |
20130329988 | COMPLEX-OBJECT DETECTION USING A CASCADE OF CLASSIFIERS - Complex-object detection using a cascade of classifiers for identifying complex-objects parts in an image in which successive classifiers process pixel patches on condition that respective discriminatory features sets of previous classifiers have been identified and selecting additional pixel patches from a query image by applying known positional relationships between an identified complex-object part and another part to be identified. | 12-12-2013 |
20130336579 | Methods for Efficient Classifier Training for Accurate Object Recognition in Images and Video - An object recognition system and method is provided which uses automated algorithmically determined negative training. Negative training with respect to a particular object classifier allows for more streamlined and efficient targeted negative training, enabling time and cost savings while simultaneously improving the accuracy of recognition based on the targeted negative training. | 12-19-2013 |
20130336580 | WEIGHTED FEATURE VOTING FOR CLASSIFICATION USING A GRAPH LATTICE - A system and method classify a test image. At least one processor receives a data graph computed from the test image. Further, a graph lattice is received. The graph lattice includes a plurality of nodes, each including a subgraph, a weight and at least one mapping of the subgraph to data graphs of a plurality of training images. The training images correspond to a plurality of classes. The data graph of the test image is mapped by the subgraphs of the nodes. Mappings between the graph lattice and the data graphs of the training images are compared with mappings between the graph lattice and the data graph of the test image to determine weighted votes of similarity between the data graphs of the training images and the data graph of the test image. The class of the test image is determined from the weighted votes. | 12-19-2013 |
20130343642 | MACHINE-LEARNT PERSON RE-IDENTIFICATION - Automated person re-identification may be assisted by consideration of attributes of the person in a joint classification with matching of the person. By both solving for similarities in a plurality of attributes and identities, discriminative interactions may be captured. Automated person re-identification may be assisted by consideration of a semantic color name. Rather than a color histogram, probability distributions are mapped to color terms of the semantic color name. Using other descriptors as well, similarity measures for the various descriptors are weighted and combined into a score. Either or both considerations may be used. | 12-26-2013 |
20130343643 | DETECTING FACIAL SIMILARITY BASED ON HUMAN PERCEPTION OF FACIAL SIMILARITY - Similar faces may be determined within images based on human perception of facial similarity. The user may provide an image including a query face to which the user wishes to find faces that are similar. Similar faces may be determined based on similarity information. Similarity information may be generated from information related to a human perception of facial similarity. Images that include faces determined to be similar, based on the similarity information, may be provided to the user as search result images. The user then may provide feedback to indicate the user's perception of similarity between the query face and the search result images. | 12-26-2013 |
20140003708 | OBJECT RETRIEVAL IN VIDEO DATA USING COMPLEMENTARY DETECTORS | 01-02-2014 |
20140003709 | ROAD MARKING DETECTION AND RECOGNITION | 01-02-2014 |
20140003710 | UNSUPERVISED LEARNING OF FEATURE ANOMALIES FOR A VIDEO SURVEILLANCE SYSTEM | 01-02-2014 |
20140016859 | SYSTEMS, METHODS, AND MEDIA FOR OPTICAL RECOGNITION - Systems, methods, and media for optical recognition are provided. In some embodiments, systems for optical recognition are provided, the systems comprising: at least one hardware processor that: identifies a plurality of fixation points in optically detected data; identifies features of the plurality of fixation points; and identifies one or more characteristics of an object represented in the optically detected data. In some embodiments, methods for optical recognition are provided, the methods comprising: identifying a plurality of fixation points in optically detected data using a hardware processor; identifying features of the plurality of fixation points using the hardware processor; and identifying one or more characteristics of an object represented in the optically detected data using the hardware processor. In some embodiments, non-transitory computer-readable media containing computer-executable instructions that, when executed by a hardware processor, cause the processor to perform these methods for optical recognition are provided. | 01-16-2014 |
20140016860 | FACIAL ANALYSIS TO DETECT ASYMMETRIC EXPRESSIONS - A system and method for facial analysis to detect asymmetric expressions is disclosed. A series of facial images is collected, and an image from the series of images is evaluated with a classifier. The image is then flipped to create a flipped image. Then, the flipped image is evaluated with the classifier. The results of the evaluation of original image and the flipped image are compared. Asymmetric features such as a wink, a raised eyebrow, a smirk, or a wince are identified. These asymmetric features are associated with mental states such as skepticism, contempt, condescension, repugnance, disgust, disbelief, cynicism, pessimism, doubt, suspicion, and distrust. | 01-16-2014 |
20140029839 | METRIC LEARNING FOR NEAREST CLASS MEAN CLASSIFIERS - A classification system and method enable improvements to classification with nearest class mean classifiers by computing a comparison measure between a multidimensional representation of a new sample and a respective multidimensional class representation embedded into a space of lower dimensionality than that of the multidimensional representations. The embedding is performed with a projection that has been learned on labeled samples to optimize classification with respect to multidimensional class representations for classes which may be the same or different from those used subsequently for classification. Each multidimensional class representation is computed as a function of a set of multidimensional representations of labeled samples, each labeled with the respective class. A class is assigned to the new sample based on the computed comparison measures. | 01-30-2014 |
20140029840 | HIGH ACCURACY LEARNING BY BOOSTING WEAK LEARNERS - A system, apparatus, method, and computer-readable medium for optimizing classifiers are disclosed. The optimization process can include receiving one or more training examples. The optimization process can further include assigning a loss parameter to each training example. The optimization process can further include optimizing each loss parameter of each training sample based on a sample variance of each training example using a non-linear function. The optimization process can further include estimating a classifier from the one or more weighted training samples. The optimization process can further include assigning a loss parameter to the classifier based on a number of training examples that the classifier correctly classified and a number of training examples that the classifier incorrectly classified. The optimization process can further include adding the weighted classifier to an overall classifier. | 01-30-2014 |
20140037196 | Systems And Methods For Spectral Authentication Of A Feature Of A Document - Systems and methods for authenticating a document are provided. In one embodiment, a method for authenticating a feature of a document includes capturing a first image of a region of a document while the region is subjected to a first wavelength of electromagnetic radiation. The region includes at least a portion of the document. The method also includes determining a first intensity value associated with the first image of the region, and comparing the first intensity value with a first training intensity value of a region of a training document. The first training intensity value is obtained using the first wavelength of electromagnetic radiation. The method also includes determining whether the document is authentic at least partially based on the comparison between the first intensity value and the first training intensity value. | 02-06-2014 |
20140037197 | METHOD FOR EDITING A MULTI-POINT FACIES SIMULATION - A computer system and a hybrid method for combining multipoint statistic and object-based methods include creating a multi-point statistics (MPS) model using a MPS method that satisfies conditioning data and constraints, the multi-point statistics being derived from a training image created using training-image generation parameters; generating one or more object-shape templates and depositional coordinates of each facies type using the parameters; positioning the templates within the MPS model such that the templates maximally correlate to the MPS model; assigning to each of the positioned templates a unique event; determining which cells are available for editing; and assigning the cells that are available for editing to facies if the cells are contained by a facies template positioned within the MPS model at its optimally correlating location. | 02-06-2014 |
20140037198 | Image Segmentation Using Hierarchical Unsupervised Segmentation and Hierarchical Classifiers - An image segmentation method includes generating a hierarchy of regions by unsupervised segmentation of an input image. Each region is described with a respective region feature vector representative of the region. Hierarchical structures are identified, each including a parent region and its respective child regions in the hierarchy. Each hierarchical structure is described with a respective hierarchical feature vector that is based on the region feature vectors of the respective parent and child regions. The hierarchical structures are classified according to a set of predefined classes with a hierarchical classifier component that is trained with hierarchical feature vectors of hierarchical structures of training images. The training images have semantic regions labeled according to the set of predefined classes. The input image is segmented into a plurality of semantic regions based on the classification of the hierarchical structures and optionally also on classification of the individual regions. | 02-06-2014 |
20140037199 | SYSTEM AND METHOD FOR DESIGNING OF DICTIONARIES FOR SPARSE REPRESENTATION - A signal processing system adapted for sparse representation of signals is provided, comprising: (i) one or more training signals; (ii) a dictionary containing signal-atoms; (iii) a representation of each training signal using a linear combination of said dictionary's signal-atoms; (iv) means for updating the representation of the training signal; (v) means for updating the dictionary one group of atoms at a time, wherein each atom update may include all representations referring to said updated atom; and (vi) means for iterating (iv) and (v) until a stopping rule is fulfilled. The system uses the K-SVD algorithm for designing dictionaries for sparse representation of signals. | 02-06-2014 |
20140044348 | IMAGE QUALITY ASSESSMENT - This disclosure concerns image quality assessment. In particular, there is described a computer implemented method, software, and computer for assessing the quality of an image. For example but not limited to, quality of the image of a face indicates the suitability of the image for use in face recognition. The invention comprises determining ( | 02-13-2014 |
20140050391 | IMAGE SEGMENTATION FOR LARGE-SCALE FINE-GRAINED RECOGNITION - A method for fine-grained image classification on an image includes automatically segmenting one or more objects of interest prior to classification; and combining segmented and original image features before performing final classification. | 02-20-2014 |
20140050392 | METHOD AND APPARATUS FOR DETECTING AND TRACKING LIPS - Provided is a method of detecting and tracking lips accurately despite a change in a head pose. A plurality of lips rough models and a plurality of lips precision models may be provided, among which a lips rough model corresponding to a head pose may be selected, such that lips may be detected by the selected lips rough model, a lips precision model having a lip shape most similar to the detected lips may be selected, and the lips may be detected accurately using the lips precision model. | 02-20-2014 |
20140056511 | SYSTEMS AND METHODS FOR CREATING A VISUAL VOCABULARY - Systems and methods for generating a visual vocabulary build a plurality of visual words via unsupervised learning on set of features of a given type; decompose one or more visual words to a collection of lower-dimensional buckets; generate labeled image representations based on the collection of lower dimensional buckets and labeled images, wherein labels associated with an image are associated with a respective representation of the image; and iteratively select a sub-collection of buckets from the collection of lower-dimensional buckets based on the labeled image representations, wherein bucket selection during any iteration after an initial iteration is based at least in part on feedback from previously selected buckets. | 02-27-2014 |
20140064609 | SENSORY INPUT PROCESSING APPARATUS AND METHODS - Sensory input processing apparatus and methods useful for adaptive encoding and decoding of features. In one embodiment, the apparatus receives an input frame having a representation of the object feature, generates a sequence of sub-frames that are displaced from one another (and correspond to different areas within the frame), and encodes the sub-frame sequence into groups of pulses. The patterns of pulses are directed via transmission channels to detection apparatus configured to generate an output pulse upon detecting a predetermined pattern within received groups of pulses that is associated with the feature. Upon detecting a particular pattern, the detection apparatus provides feedback to the displacement module in order to optimize sub-frame displacement for detecting the feature of interest. In another embodiment, the detections apparatus elevates its sensitivity (and/or channel characteristics) to that particular pulse pattern when processing subsequent pulse group inputs, thereby increasing the likelihood of feature detection. | 03-06-2014 |
20140072206 | SEMANTIC REPRESENTATION MODULE OF A MACHINE LEARNING ENGINE IN A VIDEO ANALYSIS SYSTEM - A machine-learning engine is disclosed that is configured to recognize and learn behaviors, as well as to identify and distinguish between normal and abnormal behavior within a scene, by analyzing movements and/or activities (or absence of such) over time. The machine-learning engine may be configured to evaluate a sequence of primitive events and associated kinematic data generated for an object depicted in a sequence of video frames and a related vector representation. The vector representation is generated from a primitive event symbol stream and a phase space symbol stream, and the streams describe actions of the objects depicted in the sequence of video frames. | 03-13-2014 |
20140072207 | METHOD AND SYSTEM FOR ALIGNING AND CLASSIFYING IMAGES - In one embodiment, L dimensional images are trained, mapped, and aligned to an M dimensional topology to obtain azimuthal angles. The aligned L dimensional images are then trained and mapped to an N dimensional topology to obtain 2 | 03-13-2014 |
20140072208 | SYSTEM AND METHOD FOR AUTOMATED OBJECT DETECTION IN AN IMAGE - A contour/shape detection model may use relatively simple and efficient kernels to detect target edges in an object within an image or video. A co-occurrence probability may be calculated for two or more edge features in an image or video using an object definition. Edge features may be differentiated between in response to measured contextual support, and prominent edge features may be extracted based on the measured contextual support. The object may then be identified based on the extracted prominent edge features. | 03-13-2014 |
20140079314 | Method and Apparatus for Improved Training of Object Detecting System - An adequate solution for computer vision applications is arrived at more efficiently and, with more automation, enables users with limited or no special image processing and pattern recognition knowledge to create reliable vision systems for their applications. Computer rendering of CAD models is used to automate the dataset acquisition process and labeling process. In order to speed up the training data preparation while maintaining the data quality, a number of processed samples are generated from one or a few seed images. | 03-20-2014 |
20140079315 | METHODS AND SYSTEMS FOR REDUCING MEMORY FOOTPRINTS ASSOCIATED WITH CLASSIFIERS - Methods and systems for reducing the required footprint of SNoW-based classifiers via optimization of classifier features. A compression technique involves two training cycles. The first cycle proceeds normally and the classifier weights from this cycle are used to rank the Successive Mean Quantization Transform (SMQT) features using several criteria. The top N (out of 512 features) are then chosen and the training cycle is repeated using only the top N features. It has been found that OCR accuracy is maintained using only 60 out of 512 features leading to an 88% reduction in RAM utilization at runtime. This coupled with a packing of the weights from doubles to single byte integers added a further 8× reduction in RAM footprint or a reduction of 68× over the baseline SNoW method. | 03-20-2014 |
20140079316 | SEGMENTATION CO-CLUSTERING - An approach to segmentation or clustering of a set of elements combines separate procedures and uses training data for those procedures on labeled data. This approach is applied to elements being components of an image of text (e.g., printed or handwritten). In some examples, the elements are connected sets of pixels. In images of text, the clusters can correspond to individual lines. The approach provides improved clustering performance as compared to any one of the procedures taken alone. | 03-20-2014 |
20140086481 | Systems and Methods for Visual Object Matching - Systems and methods for improving visual object recognition by analyzing query images are disclosed. In one example, a visual object recognition module may determine query images matching objects of a training corpus utilized by the module. Matched query images may be added to the training corpus as training images of a matched object to expand the recognition of the object by the module. In another example, relevant candidate image corpora from a pool of image data may be automatically selected by matching the candidate image corpora against user query images. Selected image corpora may be added to a training corpus to improve recognition coverage. In yet another example, objects unknown to a visual object recognition module may be discovered by clustering query images. Clusters of similar query images may be annotated and added into a training corpus to improve recognition coverage. | 03-27-2014 |
20140093160 | 3D Object Tracking in Multiple 2D Sequences - A tumor is tracked in multiple sequences of images acquired concurrently from different viewpoints. Features are extracted in each set of current images using a window. A regression function, subject to motion constraints, is applied to the features to obtain 3D motion parameters, which are applied to the tumor as observed in the images to obtain a 3D location of the object. Then, the shape of the 3D object at the 3D location is projected onto each image to update the location of the window for the next set of images to be processed. | 04-03-2014 |
20140093161 | CHARACTER RECOGNITION APPARATUS, CHARACTER RECOGNITION METHOD, AND COMPUTER-READABLE MEDIUM - A character recognition apparatus includes an evaluation-value output unit, a generation unit, a learning unit, and a determination unit. The evaluation-value output unit outputs evaluation values for each of different character recognition programs. Each evaluation value indicates a degree to which an inputted character pattern corresponds to each of character codes to be recognized using the character recognition program. The generation unit generates feature information for the character pattern. The feature information includes, as elements, the evaluation values. The learning unit learns classifications for feature information on a character-code-by-character-code basis based on feature information generated for a character pattern for which a character code is specified in advance. The determination unit determines a character code of an unknown character pattern whose character code is unknown, based on which classification among the learned classifications includes feature information generated for the unknown character pattern. | 04-03-2014 |
20140099020 | METHOD OF DETECTING SMOKE OF FOREST FIRE USING SPATIOTEMPORAL BOF OF SMOKE AND RANDOM FOREST - A method of detecting the smoke of a forest fire using the spatiotemporal Bag-of-Features (BoF) of the smoke and a random forest is provided. In the method, whenever each frame of a video sequence is input, a difference between the input frame and a previous frame is detected, and the input frame is set as a key frame if the difference exceeds a predetermined first threshold value. One or more moving blocks are detected in the set key frame. One or more candidate smoke blocks are extracted from the moving blocks using a smoke color model. BoF representations are generated from the detected candidate smoke blocks. Whether smoke of the candidate smoke blocks is actual smoke is determined by performing random forest learning on the generated BoF representation. | 04-10-2014 |
20140105487 | IMAGE PROCESSING DEVICE, INFORMATION GENERATION DEVICE, IMAGE PROCESSING METHOD, INFORMATION GENERATION METHOD, AND COMPUTER READABLE MEDIUM - A feature value extraction section extracts a feature value from a pixel or a group of pixels of a sampling point for every plurality of sampling points for a reference point with respect to a region point on an image, and extracts a group of feature values with respect to the reference point; the location information identification section references an LRF function indicating a correspondence of the group of feature values with respect to the reference point and the location information indicating a relative location of the region point with respect to the reference point to identify the location information corresponding to the group of feature values extracted by the feature value extraction section, and the region point identification section assumes the location indicated by the location information identified by the location information identification section as a region point of the object. | 04-17-2014 |
20140112576 | Systems and Methods for Matching Visual Object Components - Systems and methods for modeling the occurrence of common image components (e.g., sub-regions) in order to improve visual object recognition are disclosed. In one example, a query image may be matched to a training image of an object. A matched region within the training image to which the query image matches may be determined and a determination may be made whether the matched region is located within an annotated image component of the training image. When the matched region matches only to the image component, an annotation associated with the component may be identified. In another example, sub-regions within a plurality of training image corpora may be annotated as common image components including associated information (e.g., metadata). Matching sub-regions appearing in many training images of objects may be down-weighted in the matching process to reduce possible false matches to query images including common image components. | 04-24-2014 |
20140112577 | Method and Apparatus for Spawning Specialist Belief Propagation Networks - A method and apparatus for processing image data is provided. The method includes the steps of employing a main processing network for classifying one or more features of the image data, employing a monitor processing network for determining one or more confusing classifications of the image data, and spawning a specialist processing network to process image data associated with the one or more confusing classifications. | 04-24-2014 |
20140119640 | SCENARIO-SPECIFIC BODY-PART TRACKING - A human subject is tracked within a scene of an observed depth image supplied to a general-purpose body-part tracker. The general-purpose body-part tracker is retrained for a specific scenario. The general-purpose body-part tracker was previously trained using supervised machine learning to identify one or more general-purpose parameters to be used by the general-purpose body-part tracker to track a human subject. During a retraining phase, scenario data is received that represents a human training-subject performing an action specific to a particular scenario. One or more special-purpose parameters are identified from the processed scenario data. The special-purpose parameters are selectively used to augment or replace one or more general-purpose parameters if the general-purpose body-part tracker is used to track a human subject performing the action specific to the particular scenario. | 05-01-2014 |
20140119641 | CHARACTER RECOGNITION APPARATUS, CHARACTER RECOGNITION METHOD, AND COMPUTER-READABLE MEDIUM - A character recognition apparatus includes an extracting unit extracting a feature point for a line in a handwritten character, first and second generation units, a learning unit, and a determination unit. The first generation unit generates first feature data from feature points for lines including an in-same-character line (first line) and being selected from lines in character-code-specified handwritten characters (known lines). The second generation unit generates second feature data from feature points for lines including an after-character-transition line (second line) and being selected from known lines. The learning unit causes a discriminator to learn classifications for first and second lines based on the first and second feature data. The determination unit determines whether each line in character-code-unknown handwritten characters is a first or second line, based on which classification is determined by the discriminator for feature data for the line. | 05-01-2014 |
20140119642 | Segmenting Human Hairs and Faces - Systems for segmenting human hairs and faces in color images are disclosed, with methods and processes for making and using the same. The image may be cropped around the face area and roughly centered. Optionally, the illumination environment of the input image may be determined. If the image is taken under dark environment or the contrast between the face and hair regions and background is low, an extra image enhancement may be applied. Sub-processes for identifying the pose angle and chin contours may be performed. A preliminary mask for the face by using multiple cues, such as skin color, pose angle, face shape and contour information can be represented. An initial hair mask by using the abovementioned multiple cues plus texture and hair shape information may be created. The preliminary face and hair masks are globally refined using multiple techniques. | 05-01-2014 |
20140126808 | RECURSIVE CONDITIONAL MEANS IMAGE DENOISING - Methods and composition for denoising digital camera images are provided herein. The method is based on directly measuring the local statistical structure of natural images in a large training set that has been corrupted with noise mimicking digital camera noise. The measured statistics are conditional means of the ground truth pixel value given a local context of input pixels. Each conditional mean is the Bayes optimal (minimum mean squared error) estimate given the specific local context. The conditional means are measured and applied recursively (e.g., the second conditional mean is measured after denoising with the first conditional mean). Each local context vector consists of only three variables, and hence the conditional means can be measured directly without prior assumptions about the underlying probability distributions, and they can be stored in fixed lookup tables. | 05-08-2014 |
20140133742 | Detector Evolution With Multi-Order Contextual Co-Occurrence - Aspects of the present invention comprise generating and using Multi-Order Contextual co-Occurrence (MOCO) descriptors to implicitly model the high level context using detection responses from a baseline object detector. In embodiments, a 1 | 05-15-2014 |
20140133743 | Method, Apparatus and Computer Readable Recording Medium for Detecting a Location of a Face Feature Point Using an Adaboost Learning Algorithm - The present disclosure relates to detecting the location of a face feature point using an Adaboost learning algorithm. According to some embodiments, a method for detecting a location of a face feature point comprises: (a) a step of classifying a sub-window image into a first recommended feature point candidate image and a first non-recommended feature point candidate image using first feature patterns selected by an Adaboost learning algorithm, and generating first feature point candidate location information on the first recommended feature point candidate image; and (b) a step of re-classifying said sub-window image classified into said first non-recommended feature point candidate image, into a second recommended feature point candidate image and a second non-recommended feature point candidate image using second feature patterns selected by the Adaboost learning algorithm, and generating second feature point candidate location information on the second recommended feature point recommended candidate image. | 05-15-2014 |
20140133744 | Image Adjustment - Techniques are disclosed relating to automatically adjusting images. In one embodiment, an image may be automatically adjusted based on a regression model trained with a database of raw and adjusted images. In one embodiment, an image may be automatically adjusted based on a model trained by both a database of raw and adjusted images and a small set of images adjusted by a different user. In one embodiment, an image may be automatically adjusted based on a model trained by a database of raw and adjusted images and predicted differences between a user's adjustment to a small set of images and a predicted adjustment based on the database of raw and adjusted images. | 05-15-2014 |
20140140609 | ROTATION OF AN IMAGE BASED ON IMAGE CONTENT TO CORRECT IMAGE ORIENTATION - In some implementations, a method rotates images based on image content to correct image orientation. In some implementations, a method includes obtaining one or more identifications of content depicted in an image and determining a current orientation of the content depicted in the image. The current orientation is determined based on the one or more identifications of the content. An amount of rotation for the image is determined that orients the content closer to a predetermined reference orientation than to the current orientation. The image is rotated by the determined amount. | 05-22-2014 |
20140140610 | Unsupervised Object Class Discovery via Bottom Up Multiple Class Learning - Techniques for unsupervised object class discovery via bottom-up multiple class learning are described. These techniques may include receiving multiple images containing one or more object classes. The multiple images may be analyzed to extract top saliency instances and least saliency instances. These saliency instances may be clustered to generate and/or update statistical models. The statistical models may be used to discover the one or more object classes. In some instances, the statistical models may be used to discover object classes of novel images. | 05-22-2014 |
20140140611 | IMAGE PROCESSING CIRCUIT, IMAGE PROCESSING METHOD, AND DISPLAY DEVICE USING THE SAME - An image processing circuit includes a determination unit configured to classify the input image data into a first image area being a gradation area, a second image area being a non-gradation area, or a third image area being an intermediate area between the gradation area and the non-gradation area; a first pixel interpolation unit configured to generate a first output pixel value interpolated by applying a theoretically calculated coefficient to a pixel value of the input image data; a second pixel interpolation unit configured to generate a second output pixel value interpolated by applying a coefficient obtained by learning to the pixel value of the input image data; and a mixing unit configured to output the first output pixel value, the second output pixel value, or a third output pixel value, obtained by processing the first and second output pixel values, according to a classification result of the determination unit. | 05-22-2014 |
20140140612 | USER TERMINAL DEVICE, SERVER DEVICE, SYSTEM AND METHOD FOR ASSESSING QUALITY OF MEDIA DATA - A user terminal device, a server device, a system and a method for assessing quality of media data are described. The user terminal device is used for extracting artefact features from the media data and for communicating the features to the server device which is then used for determining a quality score using the artefacts and an artefact/quality score database accessible by the server device. The score, transmitted to the user terminal device, is presented to a user from which a subjective quality score and a request for re-determination are received which the user terminal device communicates to the server device. This in turn is used for re-determining the quality score and for transmitting back the re-determined quality score wherein the quality score is re-determined using the received artefacts, the received subjective quality score and the artefact/quality score database. | 05-22-2014 |
20140153819 | Learned Piece-Wise Patch Regression for Image Enhancement - Systems and methods are provided for providing learned, piece-wise patch regression for image enhancement. In one embodiment, an image manipulation application generates training patch pairs that include training input patches and training output patches. Each training patch pair includes a respective training input patch from a training input image and a respective training output patch from a training output image. The training input image and the training output image include at least some of the same image content. The image manipulation application determines patch-pair functions from at least some of the training patch pairs. Each patch-pair function corresponds to a modification to a respective training input patch to generate a respective training output patch. The image manipulation application receives an input image generates an output image from the input image by applying at least some of the patch-pair functions based on at least some input patches of the input image. | 06-05-2014 |
20140169663 | System and Method for Video Detection and Tracking - System and method embodiments are provided to enable features and functionalities for automatically detecting and localizing the position of an object in a video frame and tracking the moving object in the video over time. One method includes detecting a plurality of objects in a video frame using a combined Histograms of Oriented Gradients (HOG) and Local Binary Pattern (LBP) algorithm, highlighting the detected objects, and tracking one of the detected objects that is selected by a user in a plurality of subsequent video frames. Also included is a user device configured to detect a plurality of objects in a video frame displayed on a display screen coupled to the user device using a combined HOG and LBP algorithm, highlight the detected objects, and track one of the detected objects that is selected by a user in a plurality of subsequent video frames on the display screen. | 06-19-2014 |
20140169664 | APPARATUS AND METHOD FOR RECOGNIZING HUMAN IN IMAGE - Disclosed herein are an apparatus and method for recognizing a human in an image. The apparatus includes a learning unit and a human recognition unit. The learning unit calculates a boundary value between a human and a non-human based on feature candidates extracted from a learning image, detects a feature candidate for which an error is minimized as the learning image is divided into the human and the non-human using the calculated boundary value, and determines the detected feature candidate to be a feature. The human recognition unit extracts a candidate image where a human may be present from an acquired image, and determines whether the candidate image corresponds to a human based on the feature that is determined by the learning unit. | 06-19-2014 |
20140177948 | Generating Training Documents - A method of generating training documents for training a classifying device comprises, with a processor, determining a number of sub-samples in a number of original documents, and creating a number of pseudo-documents from the sub-samples, the pseudo-documents comprising a portion of the number of sub-samples. A device for training a classifying device comprises a processor, and a memory communicatively coupled to the processor. The memory comprises a sampling module to, when executed by the processor, determine a number of sub-samples in a number of original documents, a pseudo-document creation module to, when executed by the processor, create a number of pseudo-documents from the sub-samples, the pseudo-documents comprising a portion of the number of sub-samples, and a training module to, when executed by the processor, train a classifying device to classify textual documents based on the pseudo-documents. | 06-26-2014 |
20140177949 | UNSUPERVISED ADAPTATION METHOD AND AUTOMATIC IMAGE CLASSIFICATION METHOD APPLYING THE SAME - An automatic image classification method applying an unsupervised adaptation method is provided. A plurality of non-manually-labeled observation data are grouped into a plurality of groups. A respectively hypothesis label is set to each of the groups according to a classifier. It is determined whether each member of the observation data in each of the groups is suitable for adjusting the classifier according to the hypothesis label, and the non-manually-labeled observation data which are determined as being suitable for adjusting the classifier are set as a plurality of adaptation data. The classifier is updated according to the hypothesis label and the adaptation data. The observation data are classified according to the updated classifier. | 06-26-2014 |
20140177950 | RECOGNITION DEVICE, METHOD, AND COMPUTER PROGRAM PRODUCT - According to an embodiment, a recognition device includes a storage unit, an acquiring unit, a first calculator, a second calculator, a determining unit, and an output unit. The storage unit stores multiple training patterns each belonging to any one of multiple categories. The acquiring unit acquires a recognition target pattern to be recognized. The first calculator calculates, for each of the categories, a distance histogram representing distribution of the number of training patterns belonging to the category with respect to distances between the recognition target pattern and the training patterns belonging to the category. The second calculator analyzes the distance histogram of each of the categories to calculate confidence of the category. The determining unit determines a category of the recognition target pattern from the multiple categories by using the confidences. The output unit outputs the category of the recognition target pattern. | 06-26-2014 |
20140185924 | Face Alignment by Explicit Shape Regression - A two-level boosted regression function is learned using shape-indexed image features and correlation-based feature selection. The regression function is learned by explicitly minimizing the alignment errors over the training data. Image features are indexed based on a previous shape estimate, and features are selected based on correlation to a random projection. The learned regression function enforces non-parametric shape constraint. | 07-03-2014 |
20140185925 | BOOSTING OBJECT DETECTION PERFORMANCE IN VIDEOS - A method and system for training a special object detector to distinguish a foreground object appearing in a sequence of frames for a target domain. The sequence of frames depicts motion of the foreground object in a non-uniform background. The foreground object is detected in a high-confidence subwindow of an initial frame of the sequence, which includes computing a measure of confidence that the high-confidence subwindow includes the foreground object and determining that the measure of confidence exceeds a specified confidence threshold. The foreground object is tracked in respective positive subwindows of subsequent frames appearing after the initial frame. The subsequent frames are within a specified short period of time. The positive subwindows are used to train the special object detector to detect the foreground object in the target domain. The positive subwindows include the subwindow of the initial frame and the respective subwindows of the subsequent frames. | 07-03-2014 |
20140185926 | Demographic Analysis of Facial Landmarks - A facial image may be annotated with the plurality of facial landmarks. These facial landmarks may be points or regions of the face that are indicative, either alone or in combination with other facial landmarks, of at least one demographic characteristic. Demographic characteristics include, for example, age, race, and/or gender. Based on the demographic characteristic being analyzed, one or more of these facial landmarks may be selected and arranged into an input vector. Then, the input vector may be compared to one or more of the training vectors. An outcome of this comparison may involve in the given facial image being classified into a category germane to the analyzed demographic characteristic (e.g., an age range or age, a racial category, and/or a gender). | 07-03-2014 |
20140198980 | IMAGE IDENTIFICATION APPARATUS, IMAGE IDENTIFICATION METHOD, AND NON-TRANSITORY COMPUTER READABLE MEDIUM - An image identification apparatus includes following components. A first generative model creation unit extracts feature information from identification-target images which belong to an identification-target category, and creates a first generative model on the basis of the feature information. A classification unit applies the first generative model to each not-identification-target image which belongs to a not-identification-target category so as to determine a probability of the not-identification-target image belonging to the identification-target category, and classifies the not-identification-target image to a corresponding one of not-identification-target groups in accordance with the probability. A second generative model creation unit that extracts feature information from not-identification-target images which belong to a corresponding one of the not-identification-target groups, and creates a second generative model of each not-identification-target group on the basis of the corresponding feature information. | 07-17-2014 |
20140205186 | Techniques for Ground-Level Photo Geolocation Using Digital Elevation - Techniques for generating cross-modality semantic classifiers and using those cross-modality semantic classifiers for ground level photo geo-location using digital elevation are provided. In one aspect, a method for generating cross-modality semantic classifiers is provided. The method includes the steps of: (a) using Geographic Information Service (GIS) data to label satellite images; (b) using the satellite images labeled with the GIS data as training data to generate semantic classifiers for a satellite modality; (c) using the GIS data to label Global Positioning System (GPS) tagged ground level photos; (d) using the GPS tagged ground level photos labeled with the GIS data as training data to generate semantic classifiers for a ground level photo modality, wherein the semantic classifiers for the satellite modality and the ground level photo modality are the cross-modality semantic classifiers. | 07-24-2014 |
20140205187 | POSE CLASSIFICATION APPARATUS AND METHOD - A pose classification apparatus is provided. The apparatus includes a first image analyzer and a second image analyzer configured to estimate a body part for each pixel of an input image including a human body, a body part decider configured to calculate reliabilities of analysis results of the first image analyzer and the second image analyzer, and configured to decide the body part for each pixel of the input image based on the calculated reliabilities, and a pose estimator configured to estimate a pose of the human body included in the input image, based on the decided body part for each pixel. | 07-24-2014 |
20140205188 | Sketch Recognition System - Handwriting interpretation tools, such as optical character recognition (OCR), have improved over the years such that OCR is a common tool in business for interpreting typed text and sometimes handwritten text. OCR does not apply well to non-text-only diagrams, such as chemical structure diagrams. A method according to an embodiment of the present invention of interpreting a human-drawn sketch includes determining a local metric indicating whether a candidate symbol belongs to a certain classification based on a set of features. The set of features includes, as a feature, scores generated from feature images of the candidate symbol. Also included is determining a joint metric of multiple candidate symbols based on their respective classifications and interpreting the sketch as a function of the local and joint metrics. Sketches can be chemical composition, biological composition, electrical schematic, mechanical, or any other science- or engineering-based diagrams for which human-drawn symbols have well-known counterparts. | 07-24-2014 |
20140219553 | METHOD FOR IMPROVING CLASSIFICATION RESULTS OF A CLASSIFIER - A method for improving classification results of a classifier including receiving classification results for a plurality of elements that have been classified by a classifier as one of a plurality of classes, constructing a graph having a plurality of nodes, each node corresponding to one of the elements, and a plurality of labels, each label corresponding to one of the classes, adding edges between nodes corresponding to related elements, adding edges between each node and each label, and using a graph cut algorithm to cut edges to a node and partition the graph into classes, the graph cut algorithm using as input the classification results for the element corresponding to that node and related elements. | 08-07-2014 |
20140219554 | PATTERN RECOGNITION APPARATUS, METHOD THEREOF, AND PROGRAM PRODUCT THEREFOR - When a feature vector is converted to a reduced vector, a converting unit samples N components of interest from the M components of the feature vector, executes the process of calculating one component of the reduced vector from the N components of interest by d times to create the d-dimensional reduced vector and, the converting unit (1) excludes the components within a predetermined distance D in the same row as the previous component of interest sampled at the previous sampling, (2) excludes the components in the same column as the previous component of interest including the component k rows apart and within the distance (D−k) from the component k rows apart, and (3) samples the component of interest of this time from the remaining components after exclusion when sampling the component of interest. | 08-07-2014 |
20140241617 | CAMERA/OBJECT POSE FROM PREDICTED COORDINATES - Camera or object pose calculation is described, for example, to relocalize a mobile camera (such as on a smart phone) in a known environment or to compute the pose of an object moving relative to a fixed camera. The pose information is useful for robotics, augmented reality, navigation and other applications. In various embodiments where camera pose is calculated, a trained machine learning system associates image elements from an image of a scene, with points in the scene's 3D world coordinate frame. In examples where the camera is fixed and the pose of an object is to be calculated, the trained machine learning system associates image elements from an image of the object with points in an object coordinate frame. In examples, the image elements may be noisy and incomplete and a pose inference engine calculates an accurate estimate of the pose. | 08-28-2014 |
20140241618 | Combining Region Based Image Classifiers - Examples disclosed herein relate to combining region based image classifiers. In one implementation, a processor measures correct classification and misclassification levels associated with a first image classifier related to a first image feature region and measures correct classification and misclassification levels associated with a second image classifier related to a second image feature region. The processor may create a combined classifier based on the first image classifier correct classification and misclassification levels and based on the second image classifier correct classification and misclassification levels such that the combined classifier is related to the first image feature region and the second image feature region. | 08-28-2014 |
20140241619 | METHOD AND APPARATUS FOR DETECTING ABNORMAL MOVEMENT - Provided are a method and apparatus for detecting an abnormal movement. The apparatus includes a feature tracing unit configured to extract features of a moving object in an input image, trace a variation in position of the extracted features according to time, and ascertain trajectories of the extracted features; a topic online learning unit configured to classify the input image in units of documents which are bundles of the trajectories, and ascertain probability distribution states of topics, which constitute the classified document, by using an online learning method which is a probabilistic topic model; and a movement pattern online learning unit configured to learn a velocity and a direction for each of the ascertained topics, and learn a movement pattern by inferring a spatiotemporal correlation between the ascertained topics. | 08-28-2014 |
20140241620 | Illumination Estimation Using Natural Scene Statistics - A method for estimating illumination of an image captured by a digital system is provided that includes computing a feature vector for the image, identifying at least one best reference illumination class for the image from a plurality of predetermined reference illumination classes using the feature vector, an illumination classifier, and predetermined classification parameters corresponding to each reference illumination class, and computing information for further processing of the image based on the at least one best reference illumination class, wherein the information is at least one selected from a group consisting of color temperature and white balance gains. | 08-28-2014 |
20140241621 | GENERATING USER INSIGHTS FROM USER IMAGES AND OTHER DATA - A method of generating user insights based on one or more user images and one or more other data is provided, including receiving one or more image files on an identifiable device or service, receiving at least one of image metadata or identifiable device or service metadata, analyzing features of the received image files, based on at least one of the received image metadata or identifiable device or service metadata and generating at least one user insight for a user associated with the identifiable device or service. Also provided is a computer readable storage medium containing program code for implementing the method. | 08-28-2014 |
20140241622 | Image Type Classifier For Improved Remote Presentation Session Compression - An invention is disclosed for classifying a graphic—e.g. as text or non-text. In embodiments, machine learning is used to generate a solution for classifying graphics of a graphic based on providing the machine learning system a plurality of graphics that are already classified. The way to determine a classification is then used by a remote presentation session server to classify tiles of frames to be transmitted to a client in a remote presentation session. The server encodes the tiles based on their classifications and transmits the encoded tiles to the client. | 08-28-2014 |
20140247977 | Method and Apparatus for Learning-Enhanced Atlas-Based Auto-Segmentation - Disclosed herein are techniques for enhancing the accuracy of atlas-based auto-segmentation (ABAS) using an automated structure classifier that was trained using a machine learning algorithm. Also disclosed is a technique for training the automated structure classifier using atlas data applied to the machine learning algorithm. | 09-04-2014 |
20140247978 | PRE-SCREENING TRAINING DATA FOR CLASSIFIERS - A system and method provide recommendations for refining training data that includes a training set of digital objects. A submitter labels the digital objects in the training set with labels, which may indicate whether the object is considered positive, neutral, or negative with respect to each of a predefined set of classes. Score vectors are computed by a trained categorizer for each digital object in the labeled training set. From the score vectors, various metrics are computed, such as a representative score vector and distances of score vectors from the representative score vector for a label group, cluster, or category of the categorizer. Based on the computed metrics, heuristics are applied and the training data is evaluated and recommendations may be made to the submitter, such as proposing that mislabeled objects are relabeled. The training data may include unlabeled digital objects, in which case, the recommendations may include suggestions for labeling the unlabeled objects. | 09-04-2014 |
20140254922 | Salient Object Detection in Images via Saliency - An input image, which may include a salient object, is received by a salient object detection and localization system. The system may be trained to detect whether the input image includes a salient object. If the system fails to detect a salient object in the input image, the system may provide the sender of the input with a null result or an indication that the input image does not contain a salient object. If the system detects a salient object in the input image, the system may localize the salient object within the input image. The system may generate an output image based at least in part on the localization of the salient object. The system may provide the sender of the input image with information pertaining to the detected salient object. | 09-11-2014 |
20140254923 | IMAGE PROCESSING AND OBJECT CLASSIFICATION - A method for classifying objects from one or more images comprising generating a trained classification process and using the trained classification process to classify objects in the images. Generating the trained classification process can include extracting features from one or more training images and clustering the features into one or more groups of features termed visual words; storing data for each of the visual words, including colour and texture information, as descriptor vectors; and generating a vocabulary tree to store clusters of visual words with common characteristics. Using the trained classification process to classify objects can include extracting features from the images and clustering the features into groups of features termed visual words; searching the vocabulary tree to determine the closest matching clusters of visual words; and classifying objects based on the closest matching clusters of visual words in the vocabulary tree. | 09-11-2014 |
20140270489 | LEARNED MID-LEVEL REPRESENTATION FOR CONTOUR AND OBJECT DETECTION - Various technologies described herein pertain to constructing mid-level sketch tokens for use in tasks, such as object detection and contour detection. Sketch patches can be extracted from binary images that comprise hand-drawn contours. The hand-drawn contours in the binary images can correspond to contours in training images. The sketch patches can be clustered to form sketch token classes. Moreover, color patches from the training images can be extracted and low-level features of the color patches can be computed. Further, a classifier that labels mid-level sketch tokens can be trained. Such training of the classifier can be through supervised learning of a mapping from the low-level features of the color patches to the sketch token classes. | 09-18-2014 |
20140270490 | Real-Time Face Detection Using Combinations of Local and Global Features - An apparatus comprises a processor configured to: input an image; detect a skin area in the image to obtain an expanded rectangular facial candidate area; detect a face in the expanded rectangular facial candidate area to obtain an initial detected facial area; subject the initial detected facial area to a false alarm removal; and output a detected facial area. | 09-18-2014 |
20140270491 | Acceleration of Linear Classifiers - In one embodiment, image detection is improved or accelerated using an approximate range query to classify images. A controller is trained on a set of training feature vectors. The training feature vectors represent an image. The feature vectors are normalized to a uniform length. The controller defines a matching space that includes the set of training feature vectors. The controller is configured to identify whether an input vector for a tested image falls within the matching space based on a range query. When the input vector falls within the matching space, the tested image substantially matches the portion of the image used to train the controller. | 09-18-2014 |
20140270492 | AUTOMATIC BUILDING ASSESSMENT - Disclosed systems and methods automatically assess buildings and structures. A device may receive one or more images of a structure, such as a building or portion of the building, and then label and extract relevant data. The device may then train a system to automatically assess other data describing similar buildings or structures based on the labeled and extracted data. After training, the device may then automatically assess new data, and the assessment results may be sent directly to a client or to an agent for review and/or processing. | 09-18-2014 |
20140270493 | ADAPTABLE CLASSIFICATION METHOD - An adaptable classification method is provided. The method performs the classification by using a classification standard having a plurality of categories. The classification standard is classified into different categories based on probability ranges. The adaptable classification method includes training a classifying device with a plurality of samples and using the trained classifying device to determine the categories of the samples to obtain classification model scores of the samples, transferring, by using logistic-like functions, the classification model scores into probability values; and adjusting parameters of logistic-like functions to iterate the training of the classifying device such that the probability values conform to value ranges corresponding to categories of the classification standard. The adaptable classification method is applicable to various classification methods based on the probability ranges, and can also retrieve a specific category from the classified categories for further classification to increase the efficacy. | 09-18-2014 |
20140270494 | COMPUTER VISION AS A SERVICE - A computer vision service includes technologies to, among other things, analyze computer vision or learning tasks requested by computer applications, select computer vision or learning algorithms to execute the requested tasks based on one or more performance capabilities of the computer vision or learning algorithms, perform the computer vision or learning tasks for the computer applications using the selected algorithms, and expose the results of performing the computer vision or learning tasks for use by the computer applications. | 09-18-2014 |
20140286568 | INFORMATION PROCESSING APPARATUS AND TRAINING METHOD - An information processing apparatus, for training a classifier that classifies local regions of an object, includes a feature amount setting unit, a selection unit, and a training unit. The feature amount setting unit sets a feature amount to be used by the classifier. The selection unit selects a local region of the object based on a predetermined selection condition based on positions for obtaining the feature amount set by the feature amount setting unit. The training unit trains the classifier using the feature amount set by the feature amount setting unit and the local region selected by the selection unit. | 09-25-2014 |
20140294291 | Image Sign Classifier - Examples disclosed herein relate to an image sign classifier. In one implementation, a processor causes a user interface to be displayed to receive information related to a target sign type in an image. The processor may train an image sign classifier based on the information to recognize the target sign type and output information related to the trained classifier. | 10-02-2014 |
20140294292 | IMAGE PROCESSING APPARATUS AND IMAGE PROCESSING METHOD - In a case where generating a training image of an object to be used to generate a dictionary to be referred to in image recognition processing of detecting the object from an input image, model information of an object to be detected is set, and a luminance image of the object and a range image are input. The luminance distribution of the surface of the object is estimated based on the luminance image and the range image, and the training image of the object is generated based on the model information and the luminance distribution. | 10-02-2014 |
20140294293 | IMAGE PROCESSING CIRCUIT AND IMAGE DETECTION DEVICE - Each second selection circuit selects, out of a plurality of evaluation values, an evaluation value being in a predetermined relative positional relation with a first evaluation value as an evaluation value outputted from a first selection circuit, and outputs the selected value. The predetermined relative positional relations are different from one another among a plurality of second selection circuits. Every time a second evaluation value is outputted from the second selection circuit corresponding to the integration circuit, the integration circuit reads a weigh value corresponding to a combination of the second evaluation value and the first evaluation, which makes a pair with the second evaluation and is outputted from the first selection circuit, from a storage circuit corresponding to the second selection circuit and integrates the read values. An addition circuit at least adds a plurality of integrated values outputted from a plurality of integration circuits, and an addition value obtained thereby becomes a probability value. | 10-02-2014 |
20140294294 | IMAGE PROCESSING APPARATUS, IMAGE PROCESSING METHOD, AND PROGRAM - There is provided an image processing apparatus including: a prediction tap selection unit which selects a pixel which is a prediction tap used for prediction operation for acquiring a pixel value of a target pixel which is a target from a second image obtained by converting a first image, from the first image; a classification unit which classifies the target pixel to any class from a plurality of classes; a tap coefficient output unit which outputs a tap coefficient of a class of the target pixel from tap coefficients, acquired by learning to minimize an error between a result of the prediction operation using a student image corresponding to the first image and a teacher image corresponding to the second image; and an operation unit which acquires a pixel value of the target pixel by performing the prediction operation using the tap coefficient and the prediction tap. | 10-02-2014 |
20140294295 | APPARATUS AND METHOD FOR VIDEO SENSOR-BASED HUMAN ACTIVITY AND FACIAL EXPRESSION MODELING AND RECOGNITION - An apparatus and method for human activity and facial expression modeling and recognition are based on feature extraction techniques from time sequential images. The human activity modeling includes determining principal components of depth and/or binary shape images of human activities extracted from video clips. Independent Component Analysis (ICA) representations are determined based on the principal components. Features are determined through Linear Discriminant Analysis (LDA) based on the ICA representations. A codebook is determined using vector quantization, Observation symbol sequences in the video clips am determined. And human activities are learned using the Hidden Markov Model (HMM) based on status transition and an observation matrix. | 10-02-2014 |
20140301635 | END-TO-END VISUAL RECOGNITION SYSTEM AND METHODS - We describe an end-to-end visual recognition system, where “end-to-end” refers to the ability of the system of performing all aspects of the system, from the construction of “maps” of scenes, or “models” of objects from training data, to the determination of the class, identity, location and other inferred parameters from test data. Our visual recognition system is capable of operating on a mobile hand-held device, such as a mobile phone, tablet or other portable device equipped with sensing and computing power. Our system employs a video based feature descriptor, and we characterize its invariance and discriminative properties. Feature selection and tracking are performed in real-time, and used to train a template-based classifier during a capture phase prompted by the user. During normal operation, the system scores objects in the field of view based on their ranking. | 10-09-2014 |
20140301636 | AUTOMATED FACIAL ACTION CODING SYSTEM - An automatic facial action coding system and method can include processing an image to identify a face in the image, to detect and align one or more facial features shown in the image, and to define one or more windows on the image. One or more distributions of pixels and color intensities can be quantified in each of the one or more windows to derive one or more two-dimensional intensity distributions of one or more colors within the window. The one or more two-dimensional intensity distributions can be processed to select image features appearing in the one or more windows and to classify one or more predefined facial actions on the face in the image. A facial action code score that includes a value indicating a relative amount of the predefined facial action occurring in the face in the image can be determined for the face in the image for each of the one or more predefined facial actions. | 10-09-2014 |
20140307956 | IMAGE LABELING USING GEODESIC FEATURES - Image labeling is described, for example, to recognize body organs in a medical image, to label body parts in a depth image of a game player, to label objects in a video of a scene. In various embodiments an automated classifier uses geodesic features of an image, and optionally other types of features, to semantically segment an image. For example, the geodesic features relate to a distance between image elements, the distance taking into account information about image content between the image elements. In some examples the automated classifier is an entangled random decision forest in which data accumulated at earlier tree levels is used to make decisions at later tree levels. In some examples the automated classifier has auto-context by comprising two or more random decision forests. In various examples parallel processing and look up procedures are used. | 10-16-2014 |
20140307957 | CLASSIFIER UPDATE DEVICE, INFORMATION PROCESSING DEVICE, AND CLASSIFIER UPDATE METHOD - One aspect of the present invention provides a classifier update device that enhances accuracy of the determination made by a classifier with respect to a specific determination target. A feature quantity update unit updates a reference value of a predetermined criterion in the classifier based on target image data including a specific type of target acquired by an image acquisition unit. | 10-16-2014 |
20140307958 | INSTANCE-WEIGHTED MIXTURE MODELING TO ENHANCE TRAINING COLLECTIONS FOR IMAGE ANNOTATION - Automatic selection of training images is enhanced using an instance-weighted mixture modeling framework called ARTEMIS. An optimization algorithm is derived that in addition to mixture parameter estimation learns instance-weights, essentially adapting to the noise associated with each example. The mechanism of hypothetical local mapping is evoked so that data in diverse mathematical forms or modalities can be cohesively treated as the system maintains tractability in optimization. Training examples are selected from top-ranked images of a likelihood-based image ranking. Experiments indicate that ARTEMIS exhibits higher resilience to noise than several baselines for large training data collection. The performance of ARTEMIS-trained image annotation system is comparable to using manually curated datasets. | 10-16-2014 |
20140307959 | METHOD AND SYSTEM OF PRE-ANALYSIS AND AUTOMATED CLASSIFICATION OF DOCUMENTS - Automatic classification of different types of documents is disclosed. An image of a form or document is captured. The document is assigned to one or more type definitions by identifying one or more objects within the image of the document. A matching model is selected via identification of the document image. In the case of multiple identifications, a profound analysis of the document type is performed—either automatically or manually. An automatic classifier may be trained with document samples of each of a plurality of document classes or document types where the types are known in advance or a system of classes may be formed automatically without a priori information about types of samples. An automatic classifier determines possible features and calculates a range of feature values and possible other feature parameters for each type or class of document. A decision tree, based on rules specified by a user, may be used for classifying documents. Processing, such as optical character recognition (OCR), may be used in the classification process. | 10-16-2014 |
20140314311 | SYSTEM AND METHOD FOR CLASSIFICATION WITH EFFECTIVE USE OF MANUAL DATA INPUT - Systems and methods are disclosed herein for classifying records, such as product records, using a machine learning algorithm. After training a classification model according to a machine learning algorithm using an initial training set, records are classified and high confidence classifications identified. Remaining classifications are submitted to a crowdsourcing forum that validates or invalidates the classifications or marks them as to unclear to evaluate. Invalidated classifications are automatically analyzed to identify one or both of classification values and categories having a high proportion of invalidated classifications. Requests are transmitted to analysts to generate training data that is added to the training set. The process of classifying records and obtaining crowdsourced validation thereof may then repeat. High confidence classifications may be identified using an accuracy model trained to relate an accuracy percentage to a confidence score output by the classification model. | 10-23-2014 |
20140321737 | COLLECTION OF MACHINE LEARNING TRAINING DATA FOR EXPRESSION RECOGNITION - Apparatus, methods, and articles of manufacture for implementing crowdsourcing pipelines that generate training examples for machine learning expression classifiers. Crowdsourcing providers actively generate images with expressions, according to cues or goals. The cues or goals may be to mimic an expression or appear in a certain way, or to “break” an existing expression recognizer. The images are collected and rated by same or different crowdsourcing providers, and the images that meet a first quality criterion are then vetted by expert(s). The vetted images are then used as positive or negative examples in training machine learning expression classifiers. | 10-30-2014 |
20140328536 | Automatic Analysis of Individual Preferences For Attractiveness - A method facilitates selection of candidate matches for an individual from a database of potential applicants. A filter is calculated for the individual by processing images of people in conjunction with the individual's preferences with respect to those images. Feature sets are calculated for the potential applicants by processing images of the potential applicants. The filter is then applied to the feature sets to select candidate matches for the individual. | 11-06-2014 |
20140334718 | HUMAN ATTRIBUTE ESTIMATION SYSTEM, HUMAN ATTRIBUTE ESTIMATION APPARATUS AND HUMAN ATTRIBUTE ESTIMATION METHOD - To provide a human attribute estimation system capable of improving estimation accuracy irrespective of an environment-dependent attribute is provided. An age/gender estimation system as a human attribute estimation system is provided with: a monitoring camera photographing a human targeted by attribute estimation and generating an image; an age/gender estimating section estimating an attribute of the human shown in the image generated by the monitoring camera using an estimation parameter; and an environment-dependent attribute specifying section specifying an environment-dependent attribute, which is an attribute dependent on an installation environment of the monitoring camera. The age/gender estimating section uses a parameter generated on the basis of learning data having an environment-dependent attribute within a predetermined distance from the environment-dependent attribute acquired by the environment-dependent attribute specifying section in an environment-dependent attribute space, as the estimation parameter. | 11-13-2014 |
20140334719 | OBJECT IDENTIFICATION DEVICE - In an object identification device, each score calculator extracts a feature quantity from the image, and calculates a score using the extracted feature quantity and a model of the specified object. The score represents a reliability that the specified object is displayed in the image. A score-vector generator generates a score vector having the scores as elements thereof. A cluster determiner determines, based on previously determined clusters in which the score vector is classifiable, one of the clusters to which the score vector belongs as a target cluster. An object identifier identifies whether the specified object is displayed in the image based on one of the identification conditions. The one of the identification conditions is previously determined for the target cluster determined by the cluster determiner. | 11-13-2014 |
20140334720 | IMAGE PROCESSING DEVICE, IMAGE PROCESSING METHOD AND COMPUTER READABLE MEDIUM - The present invention generates a highly-accurate restored image while reducing computation costs by obtaining a similar effect to when the adjacent similarity of restoration patches is taken into consideration. | 11-13-2014 |
20140348419 | Auto-Maintained Document Classification - Machines, systems and methods for maintaining a representative data set in a document classification system, the method comprising: including an initial set of seed representative data in a representative data set (RDS) implemented for a knowledge base (KB), wherein the KB is trained to classify documents provided to a document classification system based on analysis of the representative documents included in the RDS and a set of rules, wherein the seed representative data includes a balanced number of representative data across a plurality of classes; updating the RDS by adding or removing representative data from the RDS based on feedback received about accuracy of classification of one or more documents by the classification system; and retraining the KB, wherein the retraining is performed based on occurrence of one or more events. | 11-27-2014 |
20140348420 | METHOD AND SYSTEM FOR AUTOMATIC SELECTION OF ONE OR MORE IMAGE PROCESSING ALGORITHM - Disclosed is a method and system for automatic algorithm selection for image processing. The invention discloses the method and system for automatically selecting the correct algorithm(s) for a varying requirement of the image for processing. The selection of algorithm is completely automatic and guided by a plurality of machine learning approaches. The system here is configured to pre-process plurality of images for creating a training data. Next, the test image is extracted, pre-processed and matched for assessing the best possible match of algorithm for processing. | 11-27-2014 |
20140355871 | SYSTEM AND METHOD FOR STRUCTURING A LARGE SCALE OBJECT RECOGNITION ENGINE TO MAXIMIZE RECOGNITION ACCURACY AND EMULATE HUMAN VISUAL CORTEX - An object recognition system and method is provided which uses automated algorithmically determined negative training Negative training with respect to a particular object classifier allows for more streamlined and efficient targeted negative training, enabling time and cost savings while simultaneously improving the accuracy of recognition based on the targeted negative training According to certain additional aspects, embodiments of the invention relate to a system and method for structuring an object recognition engine to maximize recognition accuracy and emulate human visual cortex. | 12-04-2014 |
20140363075 | IMAGE-BASED FACETED SYSTEM AND METHOD - Disclosed herein is a system and method that facilitate searching and/or browsing of images by clustering, or grouping, the images into a set of image clusters using facets, such as without limitation visual properties or visual characteristics, of the images, and representing each image cluster by a representative image selected for the image cluster. A map-reduce based probabilistic topic model may be used to identify one or more images belonging to each image cluster and update model parameters. | 12-11-2014 |
20140363076 | ESTIMATOR TRAINING METHOD AND POSE ESTIMATING METHOD USING DEPTH IMAGE - An estimator training method and a pose estimating method using a depth image are disclosed, in which the estimator training method may train an estimator configured to estimate a pose of an object, based on an association between synthetic data and real data, and the pose estimating method may estimate the pose of the object using the trained estimator. | 12-11-2014 |
20140369597 | SYSTEM AND METHOD OF CLASSIFIER RANKING FOR INCORPORATION INTO ENHANCED MACHINE LEARNING - Systems and methods are disclosed for machine classifiers that employ enhanced machine learning. The machine classification may be automated, based on the input of human classifiers, or a combination of both. The selection of human classifiers is determined by a classifier ranking or scoring process. In addition, data generated by the ranking or scoring process can be used to train the machine classifiers to more accurately classify data. | 12-18-2014 |
20140376804 | LABEL-EMBEDDING VIEW OF ATTRIBUTE-BASED RECOGNITION - In image classification, each class of a set of classes is embedded in an attribute space where each dimension of the attribute space corresponds to a class attribute. The embedding generates a class attribute vector for each class of the set of classes. A set of parameters of a prediction function operating in the attribute space respective to a set of training images annotated with classes of the set of classes is optimized such that the prediction function with the optimized set of parameters optimally predicts the annotated classes for the set of training images. The prediction function with the optimized set of parameters is applied to an input image to generate at least one class label for the input image. The image classification does not include applying a class attribute classifier to the input image. | 12-25-2014 |
20150030237 | IMAGE RESTORATION CASCADE - Image restoration cascades are described, for example, where digital photographs containing noise are restored using a cascade formed from a plurality of layers of trained machine learning predictors connected in series. For example, noise may be from sensor noise, motion blur, dust, optical low pass filtering, chromatic aberration, compression and quantization artifacts, down sampling or other sources. For example, given a noisy image, each trained machine learning predictor produces an output image which is a restored version of the noisy input image; each trained machine learning predictor in a given internal layer of the cascade also takes input from the previous layer in the cascade. In various examples, a loss function expressing dissimilarity between input and output images of each trained machine learning predictor is directly minimized during training. In various examples, data partitioning is used to partition a training data set to facilitate generalization. | 01-29-2015 |
20150030238 | VISUAL PATTERN RECOGNITION IN AN IMAGE - A system may be configured as an image recognition machine that utilizes an image feature representation called local feature embedding (LFE). LFE enables generation of a feature vector that captures salient visual properties of an image to address both the fine-grained aspects and the coarse-grained aspects of recognizing a visual pattern depicted in the image. Configured to utilize image feature vectors with LFE, the system may implement a nearest class mean (NCM) classifier, as well as a scalable recognition algorithm with metric learning and max margin template selection. Accordingly, the system may be updated to accommodate new classes with very little added computational cost. This may have the effect of enabling the system to readily handle open-ended image classification problems. | 01-29-2015 |
20150030239 | TRAINING CLASSIFIERS FOR DEBLURRING IMAGES - A classifier training system trains a classifier for evaluating image deblurring quality using a set of scored deblurred images. In some embodiments, the classifier training system trains the classifier based on a number of sub-images extracted from the scored deblurred images. An image deblurring system applies a number of different deblurring transformations to a given blurry reference image and uses the classifier trained by the classifier training system to evaluate deblurring quality, thereby finding a highest-quality deblurred image. In some embodiments, the classifier training system trains the classifier in the frequency domain, and the image deblurring system uses the classifier trained by the classifier training system to evaluate deblurring quality in the frequency domain. In some embodiments, the image deblurring system applies the different deblurring transformations iteratively. | 01-29-2015 |
20150036921 | IMAGE COMPOSITION EVALUATING APPARATUS, INFORMATION PROCESSING APPARATUS AND METHODS THEREOF - The present invention discloses an image composition evaluating apparatus, an information processing apparatus and methods thereof. The image composition evaluating apparatus comprises: a region segmentation unit configured to segment an image into a plurality of regions; a region attribution extraction unit configured to extract at least one attribution from each of the regions; a region relationship description unit configured to describe relationships among the regions based on the extracted attributions; and a composition evaluation unit configured to evaluate the composition of the image based on the extracted attributions, the described relationships and at least one preset criterion. The present invention can evaluate more kinds of images and/or more kinds of composition problems. | 02-05-2015 |
20150036922 | Method and Apparatus for Spawning Specialist Belief Propagation Networks For Adjusting Exposure Settings - A method and apparatus for processing image data is provided. The method includes the steps of employing a main processing network for classifying one or more features of the image data, employing a monitor processing network for determining one or more confusing classifications of the image data, and spawning a specialist processing network to process image data associated with the one or more confusing classifications. | 02-05-2015 |
20150043809 | Automatic Segmentation of Articulated Structures - Disclosed herein is a framework for segmenting articulated structures. In accordance with one aspect, the framework receives a target image, a reference image, statistical shape models, local appearance models and a learned landmark detector. The framework may automatically detect first centerline landmarks along centerlines of articulated structures in the target image using the learned landmark detector. The framework may then determine a non-rigid transformation function that registers second centerline landmarks along centerlines of articulated structures in the reference image with the first centerline landmarks. Mean shapes of the statistical shape models may then be deformed to the target image space by applying the non-rigid transformation function on the mean shapes. The framework may further search for candidate points in the mean shapes using the local appearance models. The mean shapes may be fitted to the candidate points to generate a segmentation mask. | 02-12-2015 |
20150043810 | IMAGE QUALITY ANALYSIS FOR SEARCHES - Image analysis includes: determining, using one or more processors, an image quality score associated with an image, including: determining a foreground and a background in the image; calculating a set of one or more characteristic parameters of the image based on the determined foreground and background; calculating the image quality score based at least in part on the set of characteristic parameters, wherein calculating the image quality score comprises using an image quality computation model that has been pre-trained; and in response to a search query, generating a set of search results that includes a set the images, wherein inclusion of the images or ranking of the search results is based at least in part on image quality scores associated with the set of images. | 02-12-2015 |
20150055854 | LEARNING BEAUTIFUL AND UGLY VISUAL ATTRIBUTES - A method for learning visual attribute labels for images includes, from textual comments associated with a corpus of images, identifying a set of candidate textual labels that are predictive of aesthetic scores associated with images in the corpus. The candidate labels in the set are clustered into a plurality of visual attribute clusters based on similarity and each of the clusters assigned a visual attribute label. For each of the visual attribute labels, a classifier is trained using visual representations of images in the corpus and respective visual attribute labels. The visual attribute labels are evaluated, based on performance of the trained classifier. A subset of the visual attribute labels is retained, based on the evaluation. The visual attribute labels can be used in processes such as image retrieval, image labeling, and the like. | 02-26-2015 |
20150055855 | LEARNING SYSTEMS AND METHODS - A sequence of images depicting an object is captured, e.g., by a camera at a point-of-sale terminal in a retail store. The object is identified, such as by a barcode or watermark that is detected from one or more of the images. Once the object's identity is known, such information is used in training a classifier (e.g., a machine learning system) to recognize the object from others of the captured images, including images that may be degraded by blur, inferior lighting, etc. In another arrangement, such degraded images are processed to identify feature points useful in fingerprint-based identification of the object. Feature points extracted from such degraded imagery aid in fingerprint-based recognition of objects under real life circumstances, as contrasted with feature points extracted from pristine imagery (e.g., digital files containing label artwork for such objects). A great variety of other features and arrangements—some involving designing classifiers so as to combat classifier copying—are also detailed. | 02-26-2015 |
20150055856 | IMAGE CLASSIFICATION - Images are classified as photos (e.g., natural photographs) or graphics (e.g., cartoons, synthetically generated images), such that when searched (online) with a filter, an image database returns images corresponding to the filter criteria (e.g., either photos or graphics will be returned). A set of image statistics pertaining to various visual cues (e.g., color, texture, shape) are identified in classifying the images. These image statistics, combined with pre-tagged image metadata defining an image as either a graphic or a photo, may be used to train a boosting decision tree. The trained boosting decision tree may be used to classify additional images as graphics or photos based on image statistics determined for the additional images. | 02-26-2015 |
20150063686 | IMAGE RECOGNITION DEVICE, IMAGE RECOGNITION METHOD, AND RECORDING MEDIUM - A delivery device according to the present invention includes a learning image storage unit, a first recognition unit, a second recognition unit, and a registration unit. The learning image storage unit stores at least one piece of learning image. The first recognition unit recognizes a recognition target image using the learning image stored in the learning image storage unit. The second recognition unit recognizes the recognition target image with finer precision than that of the first recognition unit, when the first recognition unit fails to recognize the recognition target image. When the second recognition unit succeeds in recognizing the recognition target image, the registration unit registers information on the successfully recognized recognition target image in a storage unit. | 03-05-2015 |
20150063687 | ROBUST SUBSPACE RECOVERY VIA DUAL SPARSITY PURSUIT - A computer-implemented method of detecting a foreground data in an image sequence using a dual sparse model framework includes creating an image matrix based on a continuous image sequence and initializing three matrices: a background matrix, a foreground matrix, and a coefficient matrix. Next, a subspace recovery process is performed over multiple iterations. This process includes updating the background matrix based on the image matrix and the foreground matrix; minimizing an L−1 norm of the coefficient matrix using a first linearized soft-thresholding process; and minimizing an L−1 norm of the foreground matrix using a second linearized soft-thresholding process. Then, background images and foreground images are generated based on the background and foreground matrices, respectively. | 03-05-2015 |
20150063688 | SYSTEM AND METHOD FOR SCENE TEXT RECOGNITION - Apparatus and method for performing accurate text recognition of non-simplistic images (e.g., images with clutter backgrounds, lighting variations, font variations, non-standard perspectives, and the like) may employ a machine-learning approach to identify a discriminative feature set selected from among features computed for a plurality of irregularly positioned, sized, and/or shaped (e.g., randomly selected) image sub-regions. | 03-05-2015 |
20150071528 | CLASSIFICATION OF LAND BASED ON ANALYSIS OF REMOTELY-SENSED EARTH IMAGES - Land classification based on analysis of image data. Feature extraction techniques may be used to generate a feature stack corresponding to the image data to be classified. A user may identify training data from the image data from which a classification model may be generated using one or more machine learning techniques applied to one or more features of the image. In this regard, the classification module may in turn be used to classify pixels from the image data other than the training data. Additionally, quantifiable metrics regarding the accuracy and/or precision of the models may be provided for model evaluation and/or comparison. Additionally, the generation of models may be performed in a distributed system such that model creation and/or application may be distributed in a multi-user environment for collaborative and/or iterative approaches. | 03-12-2015 |
20150071529 | LEARNING IMAGE COLLECTION APPARATUS, LEARNING APPARATUS, AND TARGET OBJECT DETECTION APPARATUS - A disclosure describes a learning image collection apparatus includes an acquisition unit, an extraction unit, a calculation unit, and a selection unit. The acquisition unit acquires an image including a target object. The extraction unit extracts, from the image, a plurality of candidate areas being candidates for the target object. The calculation unit calculates one of a first degree of similarity, a second degree of similarity, and a third degree of similarity, the first degree of similarity being a degree of similarity between one of the candidate areas and a predetermined area, the second degree of similarity being a degree of similarity between a size of the target object and a predetermined size, the third degree of similarity being a degree of similarity between the plurality of candidate areas. The selection unit selects one of the candidate areas as a target object area including the target object. | 03-12-2015 |
20150078654 | Visual Descriptors Based Video Quality Assessment Using Outlier Model - System and method for identifying erroneous videos and assessing video quality is provided. Feature vectors are generated corresponding to a plurality of frames associated with the one or more videos. The feature vectors are subsequently subjected to anomaly detection to obtain first and second normalized path lengths and normalized anomaly measures. The first and second normalized path lengths and normalized anomaly measures are provided to a regression model to identify the erroneous video. | 03-19-2015 |
20150078655 | DEVICES, SYSTEMS, AND METHODS FOR LARGE-SCALE LINEAR DISCRIMINANT ANALYSIS OF IMAGES - Systems, devices, and methods for generating hierarchical subspace maps obtain a training set of images, wherein the images in the training set of images are each associated with at least one category in a plurality of categories; organize the images in the training set of images into a category hierarchy based on the training set of images and on the plurality of categories, wherein the category hierarchy identifies each of the categories in the plurality of categories as at least one of a parent category and child category; and generate a subspace map for each parent category based on images associated with respective child categories of the parent category, thereby generating a plurality of subspace maps. | 03-19-2015 |
20150078656 | VISUALIZING AND UPDATING LONG-TERM MEMORY PERCEPTS IN A VIDEO SURVEILLANCE SYSTEM - Techniques are disclosed for visually conveying a percept. The percept may represent information learned by a video surveillance system. A request may be received to view a percept for a specified scene. The percept may have been derived from data streams generated from a sequence of video frames depicting the specified scene captured by a video camera. A visual representation of the percept may be generated. A user interface may be configured to display the visual representation of the percept and to allow a user to view and/or modify metadata attributes with the percept. For example, the user may label a percept and set events matching the percept to always (or never) result in alert being generated for users of the video surveillance system. | 03-19-2015 |
20150093021 | TABLE RECOGNIZING METHOD AND TABLE RECOGNIZING SYSTEM - Provided is a table recognizing method, comprising: parsing and analyzing metadata information in an original fixed-layout document, and extracting basic elements on a page of the document; segmenting the basic elements, extracting segmented text lines on the page, and acquiring fragments; constructing an undirected graph with respect to each of the fragments; extracting an image on the page, detecting intersection points of horizontal lines and vertical lines, detecting an external bounding box of the intersection points, and taking whether the segmented text lines fall within the external bounding box as local relationship features; training a learning model according to the local relationship features, local features of the fragments, and neighborhood relationship features among the fragments, acquiring model parameters, and establishing a table recognizing model; and invoking the table recognizing model to perform table recognizing for the document, and acquiring a recognizing result. | 04-02-2015 |
20150098646 | LEARNING USER PREFERENCES FOR PHOTO ADJUSTMENTS - In example embodiments, systems and methods for learning and using user preferences for image adjustments are presented. In example embodiments, a new image is received. A correction parameter based on previously stored user adjustments for similar images is determined. A user style that is an adjusted version of the new image is generated by applying the correction parameter. The user style is provided on a user interface. A user adjustment is received. Based on determining that a user sample image is within a predetermined threshold of closeness to the new image, data corresponding to the user sample image is replaced with new adjustment data for the new image in a database of user sample images used to generate the correction parameter. Based on determining that no user sample images are within the predetermined threshold of closeness, new adjustment data is appended to the database used to generate the correction parameter. | 04-09-2015 |
20150110386 | Tree-based Linear Regression for Denoising - Image denoising techniques are described. In one or more implementations, a denoising result is computed by a computing device for a patch of an image. One or more partitions are located by the computing device that correspond to the denoising result and a denoising operator is obtained by the computing device that corresponds to the located one or more partitions. The obtained denoising operator is applied by the computing device to the image. | 04-23-2015 |
20150110387 | METHOD FOR BINARY CLASSIFICATION OF A QUERY IMAGE - The invention relates to a method for the training of a classifier based on weakly labeled images and for the binary classification of an image. The training of the classifier comprises the steps of automatically and iteratively determining initial regions of interest for a training set and further on refining said regions of interest and adapting the classifier onto the refined regions of interest by a classifier refinement procedure. Further on, for a query image with unknown classification, an initial region of interest is determined and refined as to maximize the probability value derived at the output of said classifier. The query image is automatically assigned a negative classification label if said probability value is lower than or equal to a predetermined first threshold. The query image is automatically assigned a positive classification label if said probability value is greater than a predetermined second threshold. | 04-23-2015 |
20150110388 | SEMANTIC REPRESENTATION MODULE OF A MACHINE-LEARNING ENGINE IN A VIDEO ANALYSIS SYSTEM - A machine-learning engine is disclosed that is configured to recognize and learn behaviors, as well as to identify and distinguish between normal and abnormal behavior within a scene, by analyzing movements and/or activities (or absence of such) over time. The machine-learning engine may be configured to evaluate a sequence of primitive events and associated kinematic data generated for an object depicted in a sequence of video frames and a related vector representation. The vector representation is generated from a primitive event symbol stream and a phase space symbol stream, and the streams describe actions of the objects depicted in the sequence of video frames. | 04-23-2015 |
20150117761 | IMAGE PROCESSING METHOD AND IMAGE PROCESSING APPARATUS USING THE SAME - The invention discloses an image processing method and an imager processing apparatus using the same. The method includes the following steps: receiving an training image; finding a minimum difference among the differences; determining whether the minimum difference is larger than a first threshold; if no, generating a first output value according to the first pixel, the background candidates and a plurality of weightings corresponding to the background candidates; updating a first background candidate corresponding to the minimum difference; updating a first weighting related to the first background candidate; if yes, adding the first pixel as a new background candidate to the background candidates and adding a new weighting corresponding to the new background candidate to the weightings; and detecting whether a moving object existing in an incoming image according to the background candidates and the weightings. | 04-30-2015 |
20150117762 | Image Processing System and Image Processing Method - Provided is an image processing device that can bring about the sufficient resemblance between an original image and a restored image obtained corresponding to a low resolution input image. The image processing device includes a means that uses a dictionary for storing data associating deteriorated patches which are from a deteriorated image formed by deteriorating a prescribed image, and restoration patches which are from the prescribed image, and calculates, as a degree-of-similarity between plural input patches generated by dividing an input image and the deteriorated patches, a weighted degree-of-similarity between weighted deteriorated patches and weighted input patches, in which forms of the deteriorated patches and the input patches are reconfigured using a patch weight which is continuous weighting; a means that selects, on the basis of the weighted degree-of-similarity, a restoration patch for each input patch; and a means that combines the restoration patches to generate a restored image. | 04-30-2015 |
20150117763 | IMAGE QUALITY MEASUREMENT BASED ON LOCAL AMPLITUDE AND PHASE SPECTRA - A method and system for determining a quality metric score for image processing are described including accepting a reference image, performing a pyramid transformation on the accepted reference image to produce a predetermined number of scales, applying image division to each scale to produce reference image patches, accepting a distorted image, performing a pyramid transformation on the accepted distorted image to produce the predetermined number of scales, applying image division to each scale to produce distorted image patches, performing a local distortion calculation for corresponding reference and distorted image patches, summing local distortion calculation results for image patch pairs, multiplying results of the summation operation by a positive weight for each scale, summing the results of the multiplication operation and applying a sigmoid function to results of the second summation operation to produce the quality metric score. | 04-30-2015 |
20150117764 | EFFICIENT DISTANCE METRIC LEARNING FOR FINE-GRAINED VISUAL CATEGORIZATION - Methods and systems for distance metric learning include generating two random projection matrices of a dataset from a d-dimensional space into an m-dimensional sub-space, where m is smaller than d. An optimization problem is solved in the m-dimensional subspace to learn a distance metric based on the random projection matrices. The distance metric is recovered in the d-dimensional space. | 04-30-2015 |
20150117765 | GENERATING EVENT DEFINITIONS BASED ON SPATIAL AND RELATIONAL RELATIONSHIPS - Data from one or more sensors is input to a workflow and fragmented to produce HyperFragments. The HyperFragments of input data are processed by a plurality of Distributed Experts, who make decisions about what is included in the HyperFragments or add details relating to elements included therein, producing tagged HyperFragments, which are maintained as tuples in a Semantic Database. Algorithms are applied to process the HyperFragments to create an event definition corresponding to a specific activity. Based on related activity included in historical data and on ground truth data, the event definition is refined to produce a more accurate event definition. The resulting refined event definition can then be used with the current input data to more accurately detect when the specific activity is being carried out. | 04-30-2015 |
20150125073 | METHOD AND APPARATUS FOR PROCESSING IMAGE - A method of processing an image by using an image processing apparatus is provided. The method includes acquiring, by the image processing apparatus, a target image, extracting a shape of a target object included in the target image, determining a category including the target object based on the extracted shape, and storing the target image by mapping the target image with additional information including at least one keyword related to the category. | 05-07-2015 |
20150131898 | BLIND IMAGE DEBLURRING WITH CASCADE ARCHITECTURE - Blind image deblurring with a cascade architecture is described, for example, where photographs taken on a camera phone are deblurred in a process which revises blur estimates and estimates a blur function as a combined process. In various examples the estimates of the blur function are computed using first trained machine learning predictors arranged in a cascade architecture. In various examples a revised blur estimate is calculated at each level of the cascade using a latest deblurred version of a blurred image. In some examples the revised blur estimates are calculated using second trained machine learning predictors interleaved with the first trained machine learning predictors. | 05-14-2015 |
20150131899 | DEVICES, SYSTEMS, AND METHODS FOR LEARNING A DISCRIMINANT IMAGE REPRESENTATION - Systems, devices, and methods for generating an image representation obtain a set of low-level features from an image; generate a high-dimensional generative representation of the low-level features; generate a lower-dimensional representation of the low-level features based on the high-dimensional generative representation of the low-level features; generate classifier scores based on classifiers and on one or more of the high-dimensional generative representation and the lower-dimensional representation, wherein each classifier uses the one or more of the high-dimensional generative representation and the lower-dimensional representation as an input, and wherein each classifier is associated with a respective label; and generate a combined representation for the image based on the classifier scores and the lower-dimensional representation. | 05-14-2015 |
20150131900 | Image Cropping Using Supervised Learning - Software for supervised learning extracts a set of pixel-level features from each source image in collection of source images. Each of the source images is associated with a thumbnail created by an editor. The software also generates a collection of unique bounding boxes for each source image. And the software calculates a set of region-level features for each bounding box. Each region-level feature results from the aggregation of pixel values for one of the pixel-level features. The software learns a regression model, using the calculated region-level features and the thumbnail associated with the source image. Then the software chooses a thumbnail from a collection of unique bounding boxes in a new image, based on application of the regression model. The software uses a thumbnail received from an editor instead of the chosen thumbnail, if the chosen thumbnail is of insufficient quality as measured against a scoring threshold. | 05-14-2015 |
20150139538 | OBJECT DETECTION WITH BOOSTED EXEMPLARS - In techniques for object detection with boosted exemplars, weak classifiers of a real-adaboost technique can be learned as exemplars that are collected from example images. The exemplars are examples of an object that is detectable in image patches of an image, such as faces that are detectable in images. The weak classifiers of the real-adaboost technique can be applied to the image patches of the image, and a confidence score is determined for each of the weak classifiers as applied to an image patch of the image. The confidence score of a weak classifier is an indication of whether the object is detected in the image patch of the image based on the weak classifier. All of the confidence scores of the weak classifiers can then be summed to generate an overall object detection score that indicates whether the image patch of the image includes the object. | 05-21-2015 |
20150146973 | DISTRIBUTED SIMILARITY LEARNING FOR HIGH-DIMENSIONAL IMAGE FEATURES - A system and method for distributed similarity learning for high-dimensional image features are described. A set of data features is accessed. Subspaces from a space formed by the set of data features are determined using a set of projection matrices. Each subspace has a dimension lower than a dimension of the set of data features. Similarity functions are computed for the subspaces. Each similarity function is based on the dimension of the corresponding subspace. A linear combination of the similarity functions is performed to determine a similarity function for the set of data features. | 05-28-2015 |
20150146974 | IMAGE PROCESSING APPARATUS, IMAGE PROCESSING METHOD, AND NON-TRANSITORY COMPUTER READABLE MEDIUM - An image processing apparatus includes a first acquiring unit that acquires an image to be processed; a setting unit that sets multiple partial image areas in the image to be processed; a second acquiring unit that acquires a first classification result indicating a possibility that an object of a specific kind is included in each of the multiple partial image areas; and a generating unit that generates a second classification result indicating a possibility that the object of the specific kind is included in the image to be processed on the basis of the first classification result of each of the multiple partial image areas. | 05-28-2015 |
20150294188 | Invariant-Based Dimensional Reduction Of Object Recognition Features, Systems And Methods - A sensor data processing system and method is described. Contemplated systems and methods derive a first recognition trait of an object from a first data set that represents the object in a first environmental state. A second recognition trait of the object is then derived from a second data set that represents the object in a second environmental state. The sensor data processing systems and methods then identifies a mapping of elements of the first and second recognition traits in a new representation space. The mapping of elements satisfies a variance criterion for corresponding elements, which allows the mapping to be used for object recognition. The sensor data processing systems and methods described herein provide new object recognition techniques that are computationally efficient and can be performed in real-time by the mobile phone technology that is currently available. | 10-15-2015 |
20150294192 | MULTI-LEVEL FRAMEWORK FOR OBJECT DETECTION - The disclosure provides an approach for detecting objects in images. An object detection application receives a set of training images with object annotations. Given these training images, the object detection application generates semantic labeling for object detections, where the labeling includes lower-level subcategories and higher-level visual composites. In one embodiment, the object detection application identifies subcategories using an exemplar support vector machine (SVM) based clustering approach. Identified subcategories are used to initialize mixture components in mixture models which the object detection application trains in a latent SVM framework, thereby learning a number of subcategory classifiers that produce, for any given image, a set of candidate windows and associated subcategory labels. In addition, the object detection application learns a structured model for object detection that captures interactions among object subcategories and identifies discriminative visual composites, using subcategory labels and spatial relationships between subcategory labels to reason about object interactions. | 10-15-2015 |
20150294193 | RECOGNITION APPARATUS AND RECOGNITION METHOD - A recognition apparatus according to an embodiment of the present invention includes: a candidate region extraction unit configured to extract a subject candidate region from an image; a feature value extraction unit configured to extract a feature value related to an attribute of the image from the subject candidate region extracted by the candidate region extraction unit; an attribute determination unit configured to determine an attribute of the subject candidate region extracted by the candidate region extraction unit on the basis of the feature value extracted by the feature value extraction unit; and a determination result integration unit configured to identify an attribute of the image by integrating determination results of the attribute determination unit. | 10-15-2015 |
20150302276 | SYSTEMS AND METHODS FOR COOKWARE DETECTION - Systems and methods for cookware detection are provided. One example system includes a vision sensor positioned so as to collect a plurality of images of a cooktop. The system includes a classifier module implemented by one or more processors. The classifier module is configured to calculate a cookware score for each of the plurality of images and to use the cookware score for each of the plurality of images to classify such image as either depicting cookware or not depicting cookware. The system includes a classifier training module implemented by the one or more processors. The classifier training module is configured to train the classifier module based at least in part on a positive image training dataset and a negative image training dataset. | 10-22-2015 |
20150302566 | IMAGE PROCESSING DEVICE AND IMAGE PROCESSING METHOD - The present invention provides an image processing device whereby the probability of outputting a restored image which accurately corresponds to an original image which is included in a low-quality input image is improved. This image processing device comprises: an image group generating means for generating, from the input image, using a dictionary which stores a plurality of patch pairs wherein a degradation patch which is a patch of a degraded image wherein a prescribed image is degraded is associated with a restoration patch which is a patch of this prescribed image, a plurality of restored image candidates including a plurality of different instances of content which have a possibility of being the original content of the input image; and an image selection presentation means for clustering the generated plurality of restored image candidates, and selecting and outputting an image candidate on the basis of the result of this clustering. | 10-22-2015 |
20150310306 | ROBUST FEATURE IDENTIFICATION FOR IMAGE-BASED OBJECT RECOGNITION - Techniques are provided that include identifying robust features within a training image. Training features are generated by applying a feature detection algorithm to the training image, each training feature having a training feature location within the training image. At least a portion of the training image is transformed into a transformed image in accordance with a predefined image transformation. Transform features are generated by applying the feature detection algorithm to the transformed image, each transform feature having a transform feature location within the transformed image. The training feature locations of the training features are mapped to corresponding training feature transformed locations within the transformed image in accordance with the predefined image transformation, and a robust feature set is compiled by selecting robust features, wherein each robust feature represents a training feature having a training feature transformed location proximal to a transform feature location of one of the transform features. | 10-29-2015 |
20150310308 | METHOD AND APPARATUS FOR RECOGNIZING CLIENT FEATURE, AND STORAGE MEDIUM - In the present disclosure, a client feature and a pre-stored template image feature are obtained; the obtained client feature and template image feature are projected according to a preset projection matrix, to generate a projection feature pair, where the projection matrix is formed by training of a first template image feature of a same object and a second template image feature of a different object; and similarity calculation is performed on the projection feature pair according to a preset similarity calculation rule, to generate a similarity result and prompt the similarity result to a client. | 10-29-2015 |
20150310312 | BUSYNESS DETECTION AND NOTIFICATION METHOD AND SYSTEM - This disclosure provides a video-based method and system for busyness detection and notification. Specifically, according to an exemplary embodiment, multiple overhead image capturing devices are used to acquire video including multiple non-overlapping ROIs (regions of interest) and the video is processed to count the number of people included within the ROIs. A busyness metric is calculated based on the number of people counted and notification of the busyness metric or changes in the busyness metric is communicated to appropriate personnel, e.g., a manager of a retail store. | 10-29-2015 |
20150324632 | HEAD-POSE INVARIANT RECOGNITION OF FACIAL ATTRIBUTES - A system facilitates automatic recognition of facial expressions or other facial attributes. The system includes a data access module and an expression engine. The expression engine further includes a set of specialized expression engines, a pose detection module, and a combiner module. The data access module accesses a facial image of a head. The set of specialized expression engines generates a set of specialized expression metrics, where each specialized expression metric is an indication of a facial expression of the facial image assuming a specific orientation of the head. The pose detection module determines the orientation of the head from the facial image. Based on the determined orientation of the head and the assumed orientations of each of the specialized expression metrics, the combiner module combines the set of specialized expression metrics to determine a facial expression metric for the facial image that is substantially invariant to the head orientation. | 11-12-2015 |
20150332438 | Patch Partitions and Image Processing - Patch partition and image processing techniques are described. In one or more implementations, a system includes one or more modules implemented at least partially in hardware. The one or more modules are configured to perform operations including grouping a plurality of patches taken from a plurality of training samples of images into respective ones of a plurality of partitions, calculating an image processing operator for each of the partitions, determining distances between the plurality of partitions that describe image similarity of patches of the plurality of partitions, one to another, and configuring a database to provide the determined distance and the image processing operator to process an image in response to identification of a respective partition that corresponds to a patch taken from the image. | 11-19-2015 |
20150347855 | Clothing Stripe Detection Based on Line Segment Orientation - Examples disclosed herein relate to clothing stripe detection based on line segment orientation. A processor may determine whether a clothing region within an image includes stripes based on a stripe classifier applied to line segment information about line segments in the clothing region. The line segment information may include the number of line segments in the clothing region at each of a plurality of orientations. The processor may output information indicating the determination of whether the clothing region includes stripes | 12-03-2015 |
20150356367 | Method and Apparatus for Learning-Enhanced Atlas-Based Auto-Segmentation - Disclosed herein are techniques for enhancing the accuracy of atlas-based auto-segmentation (ABAS) using an automated structure classifier that was trained using a machine learning algorithm. Also disclosed is a technique for training the automated structure classifier using atlas data applied to the machine learning algorithm. | 12-10-2015 |
20150356376 | IMAGE PATTERN RECOGNITION SYSTEM AND METHOD - An image pattern recognition method detects a pattern in a sequence of video images or individual images from detected interest points. Feature vectors are extracted with video data from video regions around the interest points. A forest of decision trees is used to compute a set of bin values in histograms with bins corresponding to leaf nodes of the decision trees. Each bin value is a sum of contributions computed for individual interest points. Non-binary decision functions are used to compute the contributions and node dependent scale values are used to compute the arguments of the non-binary decision functions. The node dependent scale values may be computed from standard deviations of feature values found for the nodes, multiplied by a factor that is common to the nodes. This factor may be adjusted by feedback so that it can be set differently for different detection classes. | 12-10-2015 |
20150363667 | RECOGNITION DEVICE AND METHOD, AND COMPUTER PROGRAM PRODUCT - According to an embodiment, a recognition device includes a memory to store therein learning patterns each belonging to one of categories; an obtaining unit to obtain a recognition target pattern; a first calculating unit to calculate, for each category, a distance histogram representing distribution of the number of learning patterns belonging to the categories with respect to distances between the recognition target pattern and the learning patterns belonging to the categories; a second calculating unit to analyze the distance histogram of each category, and calculate a feature value of the recognition target pattern; a third calculating unit to make use of the feature value and one or more classifiers, and calculate degrees of reliability of the recognition target categories; and a determining unit to make use of the degrees of reliability and, from among the one or more recognition target categories, determine a category of the recognition target pattern. | 12-17-2015 |
20150371115 | CLASSIFICATION OF LAND BASED ON ANALYSIS OF REMOTELY-SENSED EARTH IMAGES - Land classification based on analysis of image data. Feature extraction techniques may be used to generate a feature stack corresponding to the image data to be classified. A user may identify training data from the image data from which a classification model may be generated using one or more machine learning techniques applied to one or more features of the image. In this regard, the classification module may in turn be used to classify pixels from the image data other than the training data. Additionally, quantifiable metrics regarding the accuracy and/or precision of the models may be provided for model evaluation and/or comparison. Additionally, the generation of models may be performed in a distributed system such that model creation and/or application may be distributed in a multi-user environment for collaborative and/or iterative approaches. | 12-24-2015 |
20150371397 | Object Detection with Regionlets Re-localization - An object detector includes a bottom-up object hypotheses generation unit; a top-down object search with supervised descent unit; and an object re-localization unit with a localization model. | 12-24-2015 |
20150379377 | Acceleration of Linear Classifiers - In one embodiment, image detection is improved or accelerated using an approximate range query to classify images. A controller is trained on a set of training feature vectors. The training feature vectors represent an image. The feature vectors are normalized to a uniform length. The controller defines a matching space that includes the set of training feature vectors. The controller is configured to identify whether an input vector for a tested image falls within the matching space based on a range query. When the input vector falls within the matching space, the tested image substantially matches the portion of the image used to train the controller. | 12-31-2015 |
20160004911 | RECOGNIZING SALIENT VIDEO EVENTS THROUGH LEARNING-BASED MULTIMODAL ANALYSIS OF VISUAL FEATURES AND AUDIO-BASED ANALYTICS - A computing system for recognizing salient events depicted in a video utilizes learning algorithms to detect audio and visual features of the video. The computing system identifies one or more salient events depicted in the video based on the audio and visual features. | 01-07-2016 |
20160004936 | COMPUTER VISION AS A SERVICE - A computer vision service includes technologies to, among other things, analyze computer vision or learning tasks requested by computer applications, select computer vision or learning algorithms to execute the requested tasks based on one or more performance capabilities of the computer vision or learning algorithms, perform the computer vision or learning tasks for the computer applications using the selected algorithms, and expose the results of performing the computer vision or learning tasks for use by the computer applications. | 01-07-2016 |
20160012277 | IMAGE RECOGNIZING APPARATUS, IMAGE RECOGNIZING METHOD, AND STORAGE MEDIUM | 01-14-2016 |
20160012317 | SYSTEMS, METHODS, AND DEVICES FOR IMAGE MATCHING AND OBJECT RECOGNITION IN IMAGES USING TEMPLATE IMAGE CLASSIFIERS | 01-14-2016 |
20160012318 | ADAPTIVE FEATURIZATION AS A SERVICE | 01-14-2016 |
20160019440 | Feature Interpolation - Feature interpolation techniques are described. In a training stage, features are extracted from a collection of training images and quantized into visual words. Spatial configurations of the visual words in the training images are determined and stored in a spatial configuration database. In an object detection stage, a portion of features of an image are extracted from the image and quantized into visual words. Then, a remaining portion of the features of the image are interpolated using the visual words and the spatial configurations of visual words stored in the spatial configuration database. | 01-21-2016 |
20160026897 | USING MACHINE LEARNING TO DEFINE USER CONTROLS FOR PHOTO ADJUSTMENTS - In various example embodiments, a system and method for using machine learning to define user controls for image adjustment is provided. In example embodiments, a new image to be adjusted is received. A weight is applied to reference images of a reference dataset based on a comparison of content of the new image to the reference image of the reference dataset. A plurality of basis styles is generated by applying weighted averages of adjustment parameters corresponding to the weighted reference images to the new image. Each of the plurality of basis styles comprises a version of the new image with an adjustment of at least one image control based on the weighted averages of the adjustment parameters of the reference dataset. The plurality of basis styles is provided to a user interface of a display device. | 01-28-2016 |
20160026900 | IMAGE PROCESSING DEVICE, INFORMATION STORAGE DEVICE, AND IMAGE PROCESSING METHOD - An image processing device includes an input reception section that receives a learning image and a correct answer label, a processing section that performs a process that generates classifier data and a processing target image, and a storage section. The processing section generates the processing target image that is the entirety or part of the learning image, calculates a feature quantity of the processing target image, generates the classifier data based on training data that is a set of the feature quantity and the correct answer label assigned to the learning image that corresponds to the feature quantity, generates an image group based on the learning image or the processing target image, classifies each image of the image group using the classifier data to calculate a classification score of each image, and regenerates the processing target image based on the classification score and the image group. | 01-28-2016 |
20160034786 | COMPUTERIZED MACHINE LEARNING OF INTERESTING VIDEO SECTIONS - This disclosure describes techniques for training models from video data and applying the learned models to identify desirable video data. Video data may be labeled to indicate a semantic category and/or a score indicative of desirability. The video data may be processed to extract low and high level features. A classifier and a scoring model may be trained based on the extracted features. The classifier may estimate a probability that the video data belongs to at least one of the categories in a set of semantic categories. The scoring model may determine a desirability score for the video data. New video data may be processed to extract low and high level features, and feature values may be determined based on the extracted features. The learned classifier and scoring model may be applied to the feature values to determine a desirability score associated with the new video data. | 02-04-2016 |
20160034787 | DETECTION DEVICE, LEARNING DEVICE, DETECTION METHOD, LEARNING METHOD, AND INFORMATION STORAGE DEVICE - A detection device includes an image acquisition section that acquires an image that has been captured by an imaging section, and includes an image of an object, a distance information acquisition section that acquires distance information about the distance from the imaging section to the object, a feature quantity calculation section that calculates a feature quantity from the image, a learning feature quantity storage section that stores a learning feature quantity calculated by a learning process corresponding to each of a plurality of distance ranges that are set corresponding to the distance from the imaging section to the object, and a detection section that determines a distance range that corresponds to the feature quantity based on the distance information, and detects the target area based on the learning feature quantity that corresponds to the determined distance range, and the feature quantity calculated by the feature quantity calculation section. | 02-04-2016 |
20160042254 | INFORMATION PROCESSING APPARATUS, CONTROL METHOD FOR SAME, AND STORAGE MEDIUM - A server is provided which generates learning data used in learning a classification rule for classifying an image that is input from an image input device. The server receives an initial image that is input from an image input device. The server also acquires device information on an image input device. Furthermore, the server generates an image different from the initial image using a parameter determined based on the device information to generate learning data using the generated image and the input image. | 02-11-2016 |
20160048741 | MULTI-LAYER AGGREGATION FOR OBJECT DETECTION - Object detection uses a deep or multiple layer network to learn features for detecting the object in the image. Multiple features from different layers are aggregated to train a classifier for the object. In addition or as an alternative to feature aggregation from different layers, an initial layer may have separate learnt nodes for different regions of the image to reduce the number of free parameters. The object detection is learned or a learned object detector is applied. | 02-18-2016 |
20160048742 | METHOD AND IMAGE PROCESSING APPARATUS FOR IMAGE VISIBILITY RESTORATION USING FISHER'S LINEAR DISCRIMINANT BASED DUAL DARK CHANNEL PRIOR - A method and an image processing apparatus for image visibility restoration are provided. The method includes the following steps. After an incoming hazy image is received, each of the incoming pixels is classified as either belonging to a localized light region or a non-localized light region. The localized light region is partitioned into patches according to each patch size in associated with image sizes in a training data set. Localized light patches are determined based on a FLD model and a designated patch size is accordingly determined. Adaptive chromatic parameters and dual dark channel priors corresponding to the designated patch size and a small patch size are determined. The incoming hazy image is restored according to the adaptive chromatic parameters, atmospheric light and a transmission map determined based on the dual dark channel priors to produce and output a de-hazed image. | 02-18-2016 |
20160063034 | ADDRESS RECOGNITION APPARATUS, SORTING APPARATUS, INTEGRATED ADDRESS RECOGNITION APPARATUS AND ADDRESS RECOGNITION METHOD - An address recognition apparatus has an address recognition section, and a non-addressee determination section. The address recognition section acquires, based on an image of an object, address information described on the object. The non-addressee determination section determines, based on a comparison result of information relating to first address information that is the address information acquired in the address recognition section at a desired timing, and information relating to second address information that is the address information acquired in the address recognition section before the first address information, whether or not the first address information is a non-destination address that is not an address of an addressee. | 03-03-2016 |
20160063357 | SYSTEMS AND METHODS FOR OBJECT CLASSIFICATION, OBJECT DETECTION AND MEMORY MANAGEMENT - A method for object classification by an electronic device is described. The method includes obtaining an image frame that includes an object. The method also includes determining samples from the image frame. Each of the samples represents a multidimensional feature vector. The method further includes adding the samples to a training set for the image frame. The method additionally includes pruning one or more samples from the training set to produce a pruned training set. One or more non-support vector negative samples are pruned first. One or more non-support vector positive samples are pruned second if necessary to avoid exceeding a sample number threshold. One or more support vector samples are pruned third if necessary to avoid exceeding the sample number threshold. The method also includes updating classifier model weights based on the pruned training set. | 03-03-2016 |
20160063685 | Image Denoising Using a Library of Functions - A method denoises a noisy image by, for each pixel in the noisy image, first constructing a key from a patch, wherein the patch includes locally neighboring pixels around the pixel. A function is selected from a function library using the key. Then, the function is applied to the patch to generate a corresponding noise free pixel for the pixel. | 03-03-2016 |
20160070988 | METHOD OF IMAGE ANALYSIS - A method to analyze an image and determine whether to output image data associated with an area of the image is provided. An object detection algorithm using training image data to detect an object based at least in part on a similarity of appearance of image data to data derived from the training image data is provided. Weakly detected objects are classified based on characteristics associated with the weakly detected object and may be added to the training image dataset for use in further training of the object detection algorithm. The object detection algorithm is trained with a revised dataset, the revised dataset being updated with data generated by the object detection algorithm. | 03-10-2016 |
20160078320 | Incremental Category Embedding for Categorization - There are provided systems and methods of incremental category embedding for categorization. One method including selecting one or more input categories from a plurality of input categories to be added to learned categories, determining at least one representative category from the learned categories for each input category from the one or more input categories, the at least one representative category representing the input category, and approximating the input category using the at least one representative category. | 03-17-2016 |
20160086029 | REDUCING THE SEARCH SPACE FOR RECOGNITION OF OBJECTS IN AN IMAGE BASED ON WIRELESS SIGNALS - Provided is a process including: determining that a mobile computing device has crossed a geofence associated with a merchant store; sending, to a remote classifier server, a request for object-recognition classifiers for objects in the merchant store; receiving a set of object-recognition classifiers; receiving with the mobile computing device from user a request to search for offers; capturing an image with a camera of the mobile computing device; receiving one or more wireless beacon identifiers with the mobile computing device; based on the wireless beacon identifiers, selecting a subset of the object-recognition classifiers in the set of object-recognition classifiers; and recognizing an object in the captured image based on the selected subset of the object-recognition classifiers; and requesting, from a remote offer publisher server, offers corresponding to the recognized object; and receiving offers from the remote offer publisher server; and displaying the received offers to the user. | 03-24-2016 |
20160086047 | APPARATUS AND METHOD FOR EXTRACTING FEATURE OF IMAGE INCLUDING OBJECT - At least one example embodiment discloses a method of converting a vector corresponding to an input image. The method includes receiving first-dimensional vector data associated with an input image, the input image including an object and converting the received first-dimensional vector data to second-dimensional vector data based on a projection matrix with an associated rank. A first dimension of the first-dimensional vector data is higher than a second dimension of the second-dimensional vector data. | 03-24-2016 |
20160086057 | FEATURE POINT DETECTION DEVICE, FEATURE POINT DETECTION METHOD, AND COMPUTER PROGRAM PRODUCT - According to an embodiment, a feature point detection device includes a generator to generate a K-class classifier and perform, for T times, an operation in which a first displacement vector is obtained that approximates D number of initial feature points of each training sample classified on a class-by-class basis to true feature points; a calculator to calculate, from the first displacement vectors, second displacement label vectors each unique to one second displacement vector, and a second displacement coordinate vector common to the second displacement vectors; a classifier to apply the K-class classifiers to the input image and obtain a second displacement label vector associated with a class identifier output from each K-class classifier; an adder to add up the second displacement label vectors; and a detector to detect D number of true feature points based on the initial feature points, the added label vector, and the second displacement coordinate vector. | 03-24-2016 |
20160086058 | INFORMATION PROCESSING APPARATUS AND CONTROL METHOD THEREOF - An information processing apparatus comprises: a registration unit adapted to register information required to determine at least one specific pattern in an image; an input unit adapted to input image data; a first generation unit adapted to extract a predetermined feature distribution from the input image data, and to generate a first feature distribution map indicating the feature distribution; a second generation unit adapted to generate a second feature distribution map by applying a conversion required to relax localities of the first feature distribution map to the first feature distribution map; and a determination unit adapted to determine, using sampling data on the second feature distribution map and the registered information, which of the specific patterns the image data matches. | 03-24-2016 |
20160098608 | SYSTEM AND METHOD FOR SCENE TEXT RECOGNITION - Apparatus and method for performing accurate text recognition of non-simplistic images (e.g., images with clutter backgrounds, lighting variations, font variations, non-standard perspectives, and the like) may employ a machine-learning approach to identify a discriminative feature set selected from among features computed for a plurality of irregularly positioned, sized, and/or shaped (e.g., randomly selected) image sub-regions. | 04-07-2016 |
20160098619 | EFFICIENT OBJECT DETECTION WITH PATCH-LEVEL WINDOW PROCESSING - An object detection method includes for each of a set of patches of an image, encoding features of the patch with a non-linear mapping function, and computing per-patch statistics based on the encoded features for approximating a window-level non-linear operation by a patch-level operation. Then, windows are extracted from the image, each window comprising a sub-set of the set of patches. Each of the windows is scored based on the computed patch statistics of the respective sub-set of patches. Objects, if any, can then be detected in the image, based on the window scores. The method and system allow the non-linear operations to be performed only at the patch level, reducing the computation time of the method, since there are generally many more windows than patches, while not impacting performance unduly, as compared to a system which performs non-linear operations at the window level. | 04-07-2016 |
20160104057 | TRAINING IMAGE ADJUSTMENT PREFERENCES - Some embodiments include a method of operating a computing device to learn user preferences of how to process digital images. The method can include: aggregating a user image selection and a context attribute associated therewith into a preference training database for a user, wherein the user image selection represents a record of the user's preference over at least one of adjusted versions of a base image when the adjusted versions are separately processed by different visual effects; determining a visual effect preference associated based on machine learning or statistical analysis of user image selections in the preference training database, the user image selections representing experimental records corresponding to the visual effects; updating a photo preference profile with the visual effect preference; and providing the photo preference profile to an image processor to adjust subsequently captured photographs provided to the image processor. | 04-14-2016 |
20160117553 | Method, device and system for realizing visual identification - Provided are a method, device and system for realizing visual identification. A visual identification device learns a visual identification feature of an identified device, and performs visual identification on the identified device accordingly. Thus automatic identification of the device is realized in a visual manner, and the unified manipulation on multiple identified devices can be easily realized. | 04-28-2016 |
20160117574 | Tagging Personal Photos with Deep Networks - Techniques and constructs to facilitate automatic tagging can provide improvements in image storage and searching. The constructs may enable training a deep network using tagged source images and target images. The constructs may also train a top layer of the deep network using a personal photo ontology. The constructs also may select one or more concepts from the ontology for tagging personal digital images. | 04-28-2016 |
20160125233 | SEMANTIC REPRESENTATION MODULE OF A MACHINE-LEARNING ENGINE IN A VIDEO ANALYSIS SYSTEM - A machine-learning engine is disclosed that is configured to recognize and learn behaviors, as well as to identify and distinguish between normal and abnormal behavior within a scene, by analyzing movements and/or activities (or absence of such) over time. The machine-learning engine may be configured to evaluate a sequence of primitive events and associated kinematic data generated for an object depicted in a sequence of video frames and a related vector representation. The vector representation is generated from a primitive event symbol stream and a phase space symbol stream, and the streams describe actions of the objects depicted in the sequence of video frames. | 05-05-2016 |
20160125273 | MACHINE LEARNING DEVICE AND CLASSIFICATION DEVICE FOR ACCURATELY CLASSIFYING INTO CATEGORY TO WHICH CONTENT BELONGS - An image acquisition unit of a machine learning device acquires n learning images assigned with labels to be used for categorization (n is a natural number larger than or equal to 2). A feature vector acquisition unit acquires a feature vector representing a feature from each of the n learning images. A vector conversion unit converts the feature vector for each of the n learning images to a similarity feature vector based on a similarity degree between the learning images. A classification condition learning unit learns a classification condition for categorizing the n learning images, based on the similarity feature vector converted by the vector conversion unit and the label assigned to each of the n learning images. A classification unit categorizes unlabeled testing images in accordance with the classification condition learned by the classification condition learning unit. | 05-05-2016 |
20160132751 | SYSTEM AND METHOD FOR ESTIMATING/DETERMINING THE DATE OF A PHOTO - Undated photos are organized by estimating the date of each photo. The date is estimated by building a model based on a set of reference photos having established dates, and comparing image characteristics of the undated photo to the image characteristics of the reference photos. The photo characteristics can include hues, saturation, intensity, contrast, sharpness and graininess as represented by image pixel data. Once the date of a photo is estimated, it can be tagged with identifying information, such as by using the estimated date to associate the photo with a node in a family tree. | 05-12-2016 |
20160132755 | TRAINING DATA GENERATING DEVICE, METHOD, AND PROGRAM, AND CROWD STATE RECOGNITION DEVICE, METHOD, AND PROGRAM - To provide a training data generating device capable of easily generating a large amount of training data used for machine-learning a dictionary of a discriminator for recognizing a crowd state. A person state determination unit | 05-12-2016 |
20160140389 | INFORMATION EXTRACTION SUPPORTING APPARATUS AND METHOD - According to one embodiment, an information extraction supporting apparatus includes a first acquirer, a determiner, a selector and an extractor. The first acquirer acquires a document from which at least one attribute indicating a type of desired information is extracted as an analysis target. The determiner determines whether or not the at least one attribute is valid, and obtains at least one of the valid attributes as one or more attribute candidates. The selector selects an attribute to be used for an analysis from the one or more attribute candidates as a selected attribute. The extractor extracts an expression belonging to the selected attribute from the document as an attribute expression. | 05-19-2016 |
20160140422 | IMAGE PROCESSING APPARATUS, IMAGE PROCESSING METHOD AND PROGRAM - It aims to accurately identify a class concerning classification of an object for each area of an image. A plurality of subsets are created from a plurality of learning images, and an area discriminator for performing area discrimination of the learning images is learned for each subset created. Then, area discrimination of a plurality of learning evaluation images is performed by using the plurality of learned area discriminators. Further, an integrated identifier is learned based on an area discrimination result performed to the plurality of learning evaluation images and scene information previously associated with each of the plurality of learning evaluation images. | 05-19-2016 |
20160140425 | METHOD AND APPARATUS FOR IMAGE CLASSIFICATION WITH JOINT FEATURE ADAPTATION AND CLASSIFIER LEARNING - A technique for improving the performance of image classification systems is proposed which consists of learning an adaptation architecture on top of the input features jointly with linear classifiers, e.g., SVM. This adaptation method is agnostic to the type of input feature and applies either to features built using aggregators, e.g., BoW, FV, or to features obtained from the activations or outputs from DCNN layers. The adaptation architecture may be single (shallow) or multi-layered (deep). This technique achieves a higher performance compared to current state of the art classification systems. | 05-19-2016 |
20160148076 | METHOD AND SYSTEM FOR AUTOMATING AN IMAGE REJECTION PROCESS - Systems and methods for automating an image rejection process. Features including texture, spatial structure, and image quality characteristics can be extracted from one or more images to train a classifier. Features can be calculated with respect to a test image for submission of the features to the classifier, given an operating point corresponding to a desired false positive rate. One or more inputs can be generated from the classifier as a confidence value corresponding to a likelihood of, for example: a license plate being absent in the image, the license plate being unreadable, or the license plate being obstructed. The confidence value can be compared against a threshold to determine if the image(s) should be removed from a human review pipeline, thereby reducing images requiring human review. | 05-26-2016 |
20160148077 | SYSTEMS AND METHODS FOR MACHINE LEARNING ENHANCED BY HUMAN MEASUREMENTS - In various embodiments, training objects are classified by human annotators, psychometric data characterizing the annotation of the training objects is acquired, a human-weighted loss function based at least in part on the classification data and the psychometric data is computationally derived, and one or more features of a query object are computationally classified based at least in part on the human-weighted loss function. | 05-26-2016 |
20160155016 | Method for Implementing a High-Level Image Representation for Image Analysis | 06-02-2016 |
20160155026 | CLASSIFYING METHOD, STORAGE MEDIUM, INSPECTION METHOD, AND INSPECTION APPARATUS | 06-02-2016 |
20160155027 | METHOD AND APPARATUS OF DETERMINING AIR QUALITY | 06-02-2016 |
20160162757 | MULTI-CLASS OBJECT CLASSIFYING METHOD AND SYSTEM - A multi-class object classifying method and system are disclosed herein, where the multi-class object classifying method includes the following steps: classes, first training images and second training images are received and stored, and first characteristic images and second characteristic images are respectively extracted from the first training images and the second training images; the first training images is used to generate classifiers through a linear mapping classifying method; a classifier and the second characteristic images are used to determine parameter ranges corresponding to the classes and a threshold corresponding to the classifier. When two of the parameter ranges overlap, the remaining parameter ranges except for the two overlapped parameter ranges are recorded; after another classifier is selected from the classifiers except for the classifier that has been selected, the previous steps is repeated until the parameter ranges don't overlap with each other and the parameter ranges are recorded. | 06-09-2016 |
20160162758 | Recognition Process Of An Object In A Query Image - Recognition process ( | 06-09-2016 |
20160162760 | DEVICES, SYSTEMS, AND METHODS FOR LEARNING AND IDENTIFYING VISUAL FEATURES OF MATERIALS - Devices, systems, and methods for classifying materials in a scene obtain spectral-BRDF material samples; learn feature-vector representations for the spectral-BRDF material samples based on the obtained spectral-BRDF material samples; train classifiers using the learned feature-vector representations; and generate a material classification using the trained classifiers and a new material sample. | 06-09-2016 |
20160171340 | LABELING COMPONENT PARTS OF OBJECTS AND DETECTING COMPONENT PROPERTIES IN IMAGING DATA | 06-16-2016 |
20160180198 | SYSTEM AND METHOD FOR DETERMINING CLUTTER IN AN ACQUIRED IMAGE | 06-23-2016 |
20160203386 | METHOD AND APPARATUS FOR GENERATING PHOTO-STORY BASED ON VISUAL CONTEXT ANALYSIS OF DIGITAL CONTENT | 07-14-2016 |
20160253570 | FEATURE VECTOR EXTRACTION DEVICE | 09-01-2016 |
20160253578 | LEARNING USER PREFERENCES FOR PHOTO ADJUSTMENTS | 09-01-2016 |
20160379048 | SUBSTITUTION OF HANDWRITTEN TEXT WITH A CUSTOM HANDWRITTEN FONT - Systems, apparatuses and methods may provide font substitution based on a custom font. In one example, a custom handwritten font may be generated based on a comparison between handwritten sample text and training text. In another example, handwritten original text may be converted to unique machine text based on a substitution of the handwritten original text with the custom handwritten font. Thus, a user's handwriting may be converted to the user's own best or preferred handwriting. | 12-29-2016 |
20160379068 | LATERAL SIGN PLACEMENT DETERMINATION - Systems, methods, and apparatuses are described for predicting the placement of road signs. A device receives data depicting road signs from multiple vehicles. The device analyzes a detected placement of the road signs and at least one characteristic of a collection of the data. The characteristic describes the road upon which the data was collected, an operation of the vehicle from which the data was collected, or an environment in which the data was collected. The device generates a model that associates values for the detected placement of the road signs with values for the at least one characteristic. The model may be later accessed to interpret subsequent sets of data describing one or more road signs. | 12-29-2016 |
20160379093 | DETECTION METHOD AND SYSTEM - A detection method executed by a computer, the detection method includes detecting a plurality of pupil candidates from a face image region in an image of a subject based on specific shape information, and identifying at least one pupil candidate as a pupil from among the plurality of pupil candidates based on brightness information related to an image region outside of the face image region and learning information indicating a relationship between the brightness information and a size of the pupil. | 12-29-2016 |
20170236011 | METHOD AND ARRANGEMENT FOR ASSESSING THE ROADWAY SURFACE BEING DRIVEN ON BY A VEHICLE | 08-17-2017 |
20170236029 | Identification of Individuals and/or Times Using Image Analysis | 08-17-2017 |
20170236032 | ACCURATE TAG RELEVANCE PREDICTION FOR IMAGE SEARCH | 08-17-2017 |
20170236355 | METHOD FOR SECURING AND VERIFYING A DOCUMENT | 08-17-2017 |
20180025228 | FEATURE-BASED VIDEO ANNOTATION | 01-25-2018 |
20190147620 | DETERMINING OPTIMAL CONDITIONS TO PHOTOGRAPH A POINT OF INTEREST | 05-16-2019 |
20220138470 | Techniques for Presentation Analysis Based on Audience Feedback, Reactions, and Gestures - Techniques performed by a data processing system for facilitating an online presentation session include establishing an online presentation session for conducting an online presentation for a first computing device of a presenter and a plurality of second computing devices of a plurality of participants, receiving a set of first media streams comprising presentation content from the first computing device, receiving a set of second media streams from the second computing devices of a first subset of the plurality of participants, the set of second media streams including audio content, video content, or both of first subset of the plurality of participants, analyzing the set of first media streams using one or more first machine learning models n to generate a set of first feedback results, analyzing the set of second media streams using one or more second machine learning models to identify a set of first reactions by the participants to obtain first reaction information, automatically analyzing the set of first feedback results and the set of first reactions to identify a first set of discrepancies between the set of first feedback results and the set of first reactions, and automatically updating one or more parameters of the one or more first machine learning models based on the first set of discrepancies to improve the suggestions for improving the online presentation. | 05-05-2022 |
20220138489 | METHOD OF LIVE VIDEO EVENT DETECTION BASED ON NATURAL LANGUAGE QUERIES, AND AN APPARATUS FOR THE SAME - A method of real-time video event detection includes: obtaining, based on a natural language query, a query vector; performing multimodal feature extraction on a video stream to obtain a video vector, obtaining a similarity score by comparing the query vector to the video vector; comparing the similarity score to a predetermined threshold; and activating, based on the similarity score being above the predetermined threshold, an action trigger. The multimodal feature extraction is performed using a plurality of overlapping windows that include sequential frames of the video stream. | 05-05-2022 |
20220138496 | POWER ESTIMATION USING INPUT VECTORS AND DEEP RECURRENT NEURAL NETWORKS - A method includes generating a plurality of input vectors based on input signals to an electric circuit, selecting a subset of the plurality of input vectors, and determining a plurality of datapoints based on the selected subset of the plurality of input vectors. Each datapoint of the plurality of datapoints indicates a power consumption of the electric circuit corresponding to an input vector of the selected subset of the input vectors. The method also includes generating, by a processor, a plurality of vector sequences based on the selected subset of the plurality of input vectors. Each vector sequence of the plurality of vector sequences includes a portion of the selected subset of the plurality of input vectors arranged chronologically. The method further includes training a machine learning model based on a first subset of the plurality of vector sequences and a corresponding first subset of the plurality of datapoints. | 05-05-2022 |
20220138498 | COMPRESSION SWITCHING FOR FEDERATED LEARNING - Methods for compression switching that includes distributing a model to client nodes, which use the model to generate a gradient vector (GV) based on a client node data set. The method includes receiving a model update that includes a gradient sign vector (GSV) based on the gradient vector; generating an updated model using the GSV; and distributing the updated model to the client nodes. The client node uses the updated model to generate a second GV based on a second client node data set. The method also includes a determination that a compression switch condition exists; based on the determination, transmitting an instruction to the client node to perform a compression switch; receiving, in response to the instruction, another model update including a subset GSV based on the second gradient vector; generating a second updated model using the subset GSV; and distributing the second updated model to the client nodes. | 05-05-2022 |
20220138513 | MULTI-SENSOR DATA OVERLAY FOR MACHINE LEARNING - The present invention relates to the reduction of multi-sensor data when used as input to machine-learning (ML) models. Typically, ML models use sensor data to learn characteristics of a problem domain. This data is usually input to the ML model in an end-to-end fashion: the data from sensor | 05-05-2022 |
20220138899 | METHODS AND APPARATUSES FOR PROCESSING IMAGE, METHODS AND APPARATUSES FOR TRAINING IMAGE RECOGNITION NETWORK AND METHODS AND APPARATUSES FOR RECOGNIZING IMAGE - The present disclosure relates to methods and apparatuses for processing an image, training an image recognition network and recognizing an image. The method of processing an image includes: obtaining a plurality of original images from an original image set, where at least one of the plurality of original images includes an annotation area; obtaining at least one first image by splicing the plurality of original images; for each of the at least one first image, adjusting a shape and/or size of the first image based on the plurality of original images to form a second image; obtaining respective positions of the at least one annotation area in the second image by converting respective positions of the at least one annotation area in the plurality of original images. | 05-05-2022 |
20220138984 | CONTROL METHOD, LEARNING DEVICE, DISCRIMINATION DEVICE AND PROGRAM - The learning device includes a first trainer, a candidate area determinator and a second trainer. On the basis of a training image and a training label including the correct coordinate value relating to the feature point included in the training image, the first trainer generates a first discriminator learned to output the predicted coordinate value relating to a feature point from an input image. The candidate area determinator determines a candidate area of the feature point in the training image based on the predicted coordinate value outputted by inputting the training image to the first discriminator. The second trainer generates the second discriminator, which is learned to output the reliability map indicating the reliability to the feature point at each block in an input image, on the basis of a cut-out image that is the candidate area cut out from the training image. | 05-05-2022 |
20220139070 | LEARNING APPARATUS, ESTIMATION APPARATUS, DATA GENERATION APPARATUS, LEARNING METHOD, AND COMPUTER-READABLE STORAGE MEDIUM STORING A LEARNING PROGRAM - A learning apparatus according to one or more embodiments executes, with respect to each learning data set, a first training step of training a second encoder and a second metadata identifier such that the identification result by the second metadata identifier matches the metadata, a second training step of training encoders and an estimator such that the result of estimation performed by the estimator matches correct answer data, a third training step of training a first metadata identifier such that the result of identification performed by the first metadata identifier matches the metadata, and a fourth training step of training a first encoder such that the result of identification performed by the first metadata identifier does not match the metadata. The third training step and the fourth training step are alternatingly and repeatedly executed. | 05-05-2022 |