Abstract: An introduction system is capable of identifying, with a high degree of precision, applicants who fulfill recruiter's requirements. An applicant identification unit 81 identifies applicants who satisfy minimum criteria on the basis of images input by applicants and a threshold image set for use in discriminating the minimum criteria for determining whether a recruiter is satisfied. A notification unit 82 notifies the recruiter of the applicants identified by the applicant identification unit 81. The introduction system may further include a threshold image set determination unit that determines the threshold image set on the basis of a result of a determination made by the recruiter on whether a sample image set prepared as samples fulfill the recruiter's requirements.

Abstract: 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. A partitioned modeling method receives an image region and performs feature extraction on the image region to generate characterization feature. A hierarchical partitioning is performed using the characterization feature to generate hierarchical partitions. A model generation is performed using the hierarchical partitions to generate partition model.

Abstract: A training method for a support vector machine, including executing an iterative process on a training set of data to determine parameters defining the machine, the iterative process being executed on the basis of a differentiable form of a primal optimization problem for the parameters, the problem being defined on the basis of the parameters and the data set.

Abstract: A system and process for recognizing documents by type and understanding at least a portion of the contents thereof. The process includes the steps of providing a document in electronic form, determining a set of facts, data and information about the document, providing the set of facts, data and information to a reasoning management unit, receiving formatted knowledge from a rule verification, validation and management unit, using the formatted knowledge to prepare an expert system having an inference engine, matching the facts, data and information against formatted knowledge using the inference engine and determining a set of applicable rules and executing actions of applicable rules.

Abstract: 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.

Abstract: An image analyzer for detecting a target object from image data, includes a reference detection unit, a primary detection unit, a secondary detection unit and an output unit. The reference detection unit detects a reference object included in the image data. The primary detection unit detects candidates for the target object from the image data. The secondary detection unit specifies a portion including the target object from among the candidates, by using a correlation between a feature of the detected reference object and a feature of the candidates. The output unit outputs information representing the portion including the target object specified by the secondary detection unit.

Abstract: A system and method for support vector machine plus (SVM+) computations include selecting a set of indexes for a target function to create a quadratic function depending on a number of variables, and reducing the number of variables to two in the quadratic function using linear constraints. An extreme point is computed for the quadratic function in closed form. A two-dimensional set is defined where the indexes determine whether a data point is in the two-dimensional set or not. A determination is made of whether the extreme point belongs to the two-dimensional set. If the extreme point belongs to the two-dimensional set, the extreme point defines a maximum and defines a new set of parameters for a next iteration. Otherwise, the quadratic function is restricted on at least one boundary of the two-dimensional set to create a one-dimensional quadratic function. The steps are repeated until the maximum is determined.

Abstract: 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.

Abstract: An image processing device which simultaneously secures and extracts a background image, at least two object images, a shape of each object image and motion of each object image, from among plural images, the image processing device including an image input unit (101) which accepts input of plural images; a hidden parameter estimation unit (102) which estimates a hidden parameter based on the plural images and a constraint enforcement parameter, which indicates a condition of at least one of the hidden parameters, using an iterative learning method; a constraint enforcement parameter learning unit (103) which learns a constraint enforcement parameter related to the hidden parameter using an estimation result from the hidden parameter estimation unit as a training signal; and a complementary learning unit (104) which causes the estimation of the hidden parameter and the learning of the constraint enforcement parameter, which utilize the result from the learning of the hidden parameter, to be iterated.

Abstract: Invention describes a method and system for detecting and tracking an object in a sequence of images. For each image the invention determines an object descriptor from a tracking region in a current image in a sequence of images, in which the tracking region corresponds to a location of an object in a previous image. A regression function is applied to the descriptor to determine a motion of the object from the previous image to the current image, in which the motion has a matrix Lie group structure. The location of the tracking region is updated using the motion of the object.

Abstract: A learning device includes: a generating unit configured to generate an image having different resolution from an input image; an extracting unit configured to extract a feature point serving as a processing object from an image generated by the generating unit; a calculating unit configured to calculate the feature amount of the feature point by subjecting the feature point to filter processing employing a predetermined filter; and an identifier generating unit configured to generate an identifier for detecting a predetermined target object from the image by statistical learning employing the feature amount; with the filter including a plurality of regions, and the calculating unit taking the difference value of difference within the regions as the feature amount.

Abstract: Described is a technology by which online recognition of handwritten input data is combined with offline recognition and processing to obtain a combined recognition result. In general, the combination improves overall recognition accuracy. In one aspect, online and offline recognition is separately performed to obtain online and offline character-level recognition scores for candidates (hypotheses). A statistical analysis-based combination algorithm, an AdaBoost algorithm, and/or a neural network-based combination may determine a combination function to combine the scores to produce a result set of one or more results. Online and offline radical-level recognition may be performed. For example, a HMM recognizer may generate online radical scores used to build a radical graph, which is then rescored using the offline radical recognition scores. Paths in the rescored graph are then searched to provide the combined recognition result, e.g., corresponding to the path with the highest score.

Abstract: Methods for predicting states of a subject are presented. For example, a method for predicting states of a subject includes obtaining training data comprising a plurality of variables, obtaining training states associated with the training data, and forming a predictive model according to the training data and the training states, the predictive model predictive of the training states. The forming of the predictive model includes extracting one or more hidden components from the training data. The extracting of the one or more hidden components includes regression analysis including determining one or more relationships between the one or more hidden components and the plurality of variables, and determining one or more relationships between the one or more hidden components and the training states. A number of the one or more hidden components is less than a number of the plurality of variables and greater than a number of the training states.

Abstract: An efficient, effective and at times superior object detection and/or recognition (ODR) function may be built from a set of Bayesian stumps. Bayesian stumps may be constructed for each feature and object class, and the ODR function may be constructed from the subset of Bayesian stumps that minimize Bayesian error for a particular object class. That is, Bayesian error may be utilized as a feature selection measure for the ODR function. Furthermore, Bayesian stumps may be efficiently implemented as lookup tables with entries corresponding to unequal intervals of feature histograms. Interval widths and entry values may be determined so as to minimize Bayesian error, yielding Bayesian stumps that are optimal in this respect.

Abstract: A pose estimating device includes: a pose dictionary; an image feature extracting unit configured to extract observed image feature information; a past information storing unit configured to store past pose estimating information of the articulated object; a pose predicting unit configured to predict a present pose; a node predicting unit configured to calculate a prior probability as to whether each nodes includes a present pose; an identifying unit configured to calculate a likelihood of the observed image feature information for each node; a node probability calculating unit configured to calculate a probability in which the present pose belongs to the node in the upper layer; and a pose estimating unit configured to calculate pose information.

Abstract: Methods for detecting a salient object in an input image are described. For this, the salient object in an image may be defined using a set of local, regional, and global features including multi-scale contrast, center-surround histogram, and color spatial distribution. These features are optimally combined through conditional random field learning. The learned conditional random field is then used to locate the salient object in the image. The methods can also use image segmentation, where the salient object is separated from the image background.

Abstract: Systems and methods are described for face recognition using discriminatively trained orthogonal rank one tensor projections. In an exemplary system, images are treated as tensors, rather than as conventional vectors of pixels. During runtime, the system designs visual features—embodied as tensor projections—that minimize intraclass differences between instances of the same face while maximizing interclass differences between the face and faces of different people. Tensor projections are pursued sequentially over a training set of images and take the form of a rank one tensor, i.e., the outer product of a set of vectors. An exemplary technique ensures that the tensor projections are orthogonal to one another, thereby increasing ability to generalize and discriminate image features over conventional techniques.

Abstract: A calculation processing apparatus for executing network calculations defined by hierarchically connecting a plurality of logical processing nodes that apply calculation processing to input data, sequentially designates a processing node which is to execute calculation processing based on sequence information that specifies an execution order of calculations of predetermined processing units to be executed by the plurality of processing nodes, so as to implement the network calculations, and executes the calculation processing of the designated processing node in the processing unit to obtain a calculation result. The calculation apparatus allocates partial areas of a memory to the plurality of processing nodes as ring buffers, and writes the calculation result in the memory while circulating a write destination of data to have a memory area corresponding to the amount of the calculation result of the processing unit as a unit.

Abstract: A data processing apparatus processes input data and outputs the processed data. The data processing apparatus includes a data processing section and a real-time learning section. The data processing section processes the input data by a predetermined processing method and outputs the processed data. The real-time learning section controls such that the processing method is learned in real time and the data processing section processes the input data by the learned processing method, so that the output data is improved as time elapses.

Abstract: This invention is to enable retrieving of a content the searcher imagines in mind. The search method includes: obtaining a query brain image representing a brain activity state of a searcher when perceiving or imagining the content to be retrieved; identifying the content corresponding to the query brain image by using a structure associating a content with a brain image representing the brain activity state when perceiving the content; and outputting the content corresponding to the query brain image. Thus, by using the query brain image, without specifically indicating the content such as a drawing, the searcher can extract the pertinent content only by imaging the content.

Abstract: Techniques are disclosed for visually conveying an event map. The event map may represent information learned by a surveillance system. A request may be received to view the event map for a specified scene. The event map may be generated, including a background model of the specified scene and at least one cluster providing a statistical distribution of an event in the specified scene. Each statistical distribution may be derived from data streams generated from a sequence of video frames depicting the specified scene captured by a video camera. Each event may be observed to occur at a location in the specified scene corresponding to a location of the respective cluster in the event map. The event map may be configured to allow a user to view and/or modify properties associated with each cluster. For example, the user may label a cluster and set events matching the cluster to always (or never) generate an alert.

Abstract: In a first exemplary embodiment of the present invention, an automated, computerized method is provided for determining illumination information in an image. According to a feature of the present invention, the method comprises the steps of identifying depth information in the image, identifying spatio-spectral information for the image, as a function of the depth information and utilizing the spatio-spectral information to identify illumination flux in the image.

Abstract: A conditional active shape model wherein a training set of images of objects in a class of objects to be identified, such as vascular cross-sections, is supplemented with training observations of at least one second characteristic of the object. A conditional mean shape of the objects is calculated, conditioned on the second characteristic, thereby reducing the size of the probable search space for the shape. A conditional covariance matrix of the shapes is calculated, conditioned on the second characteristic, and the eigenvectors of the conditional covariance matrix corresponding to largest eigenvalues are calculated. The conditional mean shape, and the eigenvalues and eigenvectors of the conditional covariance matrix are then used in an active shape model to identify the shapes of objects in subsequent images.

Abstract: Techniques are described to leverage a set of sample or example matched pairs of strings to learn string transformation rules, which may be used to match data records that are semantically equivalent. In one embodiment, matched pairs of input strings are accessed. For a set of matched pairs, a set of one or more string transformation rules are learned. A transformation rule may include two strings determined to be semantically equivalent. The transformation rules are used to determine whether a first and second string match each other.

Abstract: An index representing the probability that a fraction image is a face image including a face in an input image is calculated for each of the positions of the face to be detected on the basis of a feature value. When the sum of the indexes of the fraction images is not smaller than the first threshold value, the image formed by the fraction images is determined to be a face image.

Abstract: Provided is an image analyzing apparatus for efficiently performing detection of an object and tracking of a specified object, including a feature value recording section that records a plurality of reference feature values different in type from each other; a feature value extracting section that extracts a plurality of feature values different in type from each other, from each of a plurality of moving image constituent images included in a moving image; an object extracting section that extracts an object from the moving image constituent images, based on a degree of matching of the plurality of extracted feature values with respect to the plurality of reference feature values recorded in the feature value recording section; a reference feature value calculating section that calculates, from the plurality of reference feature values recorded in the feature value recording section, a plurality of reference feature values adjusted to the feature values of the extracted object, to a predetermined degree corre

Abstract: A classification system is described for resiliently classifying data. In various embodiments, the classification system constructs a combined classifier based on multiple classifiers that are constructed to classify a set of training data. The combined classifier can be constructed in parallel with the multiple classifiers and applied to classify data.

Abstract: An access system in one embodiment that first determines that someone has correct credentials by using a non-biometric authentication method such as typing in a password, presenting a Smart card containing a cryptographic secret, or having a valid digital signature. Once the credentials are authenticated, then the user must take at least two biometric tests, which can be chosen randomly. In one approach, the biometric tests need only check a template generated from the user who desires access with the stored templates matching the holder of the credentials authenticated by the non-biometric test. Access desirably will be allowed when both biometric tests are passed.

Abstract: A method and apparatus are provided for adapting an exponential probability model. In a first stage, a general-purpose background model is built from background data by determining a set of model parameters for the probability model based on a set of background data. The background model parameters are then used to define a prior model for the parameters of an adapted probability model that is adapted and more specific to an adaptation data set of interest. The adaptation data set is generally of much smaller size than the background data set. A second set of model parameters are then determined for the adapted probability model based on the set of adaptation data and the prior model.

Abstract: A method for configuring a machine learning model for use in a motion characterisation process. The method comprises, in a processing system, acquiring user characterisations for respective portions of at least one video sequence and configuring the model using the user characterisations and at least one property associated with the respective portions. Any inconsistency in the user characterisations is then determined, using the model, with an indication of the inconsistency being displayed. This allows the user to select at least one option for addressing the inconsistency, with the model being reconfigured based on the selected at least one option.

Abstract: A processor architecture for a learning machine is presented which uses a massive array of processing elements having local, recurrent connections to form global associations between functions defined on manifolds. Associations between these functions provide the basis for learning cause-and-effect relationships involving vision, audition, tactile sensation and kinetic motion. Two arbitrary input signals hold each other in place in a manifold association processor and form the basis of short-term memory.

Abstract: The present invention provides an improved system and method for object detection with histogram of oriented gradient (HOG) based support vector machine (SVM). Specifically, the system provides a computational framework to stably detect still or not moving objects over a wide range of viewpoints. The framework includes providing a sensor input of images which are received by the “focus of attention” mechanism to identify the regions in the image that potentially contain the target objects. These regions are further computed to generate hypothesized objects, specifically generating selected regions containing the target object hypothesis with respect to their positions. Thereafter, these selected regions are verified by an extended HOG-based SVM classifier to generate the detected objects.

Abstract: 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.

Abstract: A Recognition Frame presents multi-level application elements to the user simultaneously through a computer graphical user interface. The interface consists of an image display panel for displaying image channels; a data display panel for displaying object measurements and summary statistics; a configuration display panel for displaying recipe content; a master tab for selecting the panels. It also consists of a processing toolbar for context dependent processing tool display. The Recognition Frame further comprises a second side frame for data object display and charting. The second side frame has a tabular arrangement consisting of properties tab, controls tab, and charts tab. The Recognition Frame links application elements through a complex data model wherein interface display is automatically updated when one element is changed.

Abstract: A computerized method, system and computer program for the computerized fractal-based analysis of a structure as presented in a pattern on a medical image. Image data is generated from the medical image and a region of interest is selected. The image data is digitized and analyzed to reveal fractal-based computer-generated features of a texture of the image data. Then a qualifier is applied to the computer-generated features to obtain fractal characteristics of the image data. A multi-fractal nature is observed for the texture of the region of interest. A marker for assessing a risk of a disease is yielded based on the multi-fractal nature of the texture.

Abstract: Iconic images for a given object or object category may be identified in a set of candidate images by using a learned probabilistic composition model to divide each candidate image into a most probable rectangular object region and a background region, ranking the candidate images according to the maximal composition score of each image, removing non-discriminative images from the candidate images, clustering highest-ranked candidate images to form clusters, wherein each cluster includes images having similar object regions according to a feature match score, selecting a representative image from each cluster as an iconic image of the object category, and causing display of the iconic image. The composition model may be a Naïve Bayes model that computes composition scores based on appearance cues such as hue, saturation, focus, and texture. Iconic images depict an object or category as a relatively large object centered on a clean or uncluttered contrasting background.

Abstract: Systems and methods for automated pattern recognition and detection of avian influenza virus in a data set corresponding to an aspect of a biological sample. The method includes receiving a first data set corresponding to a first aspect of a first biological sample, analyzing the first data set using results of a first series of algorithms processed on a second data set corresponding to an aspect of a second biological sample known to contain avian influenza virus, generating an algorithm value cache for the first data set by running a second series of algorithms on the first data set, generating a match result by comparing the algorithm value cache with the results of the first series of algorithms, and performing a processing action based on the generated match result.

Abstract: 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.

Abstract: A system and method of machine learning that uses an inverse matrix of a reference similarity matrix as a transformation matrix. The reference similarity matrix relates a reference set of objects to themselves using a distance metric such as an image edit distance. The transformation matrix is used to improve the performance of query vectors in classifying or identifying digital representations of an unknown object. The query vector is a measure of similarity between the unknown object and the members of the reference set. Multiplying the query vector by the transformation matrix produces an improved query vector having improved similarity scores. The highest improved similarity score indicates the best match member of the reference set If the similarity score is high enough, the unknown object may either be classified as belonging to the same class, or recognized as being the same object, as the best match object.

Abstract: An image processing apparatus and method, and a program-recorded recording medium perform detecting features of image information for classes which is input for learning, calculating the degrees of belongingness of the detected features to the classes, weighting pieces of the input image information based on the degrees of belongingness, and calculating prediction coefficients for the classes based on the weighted pieces of the image information.

Abstract: Given an image of structured and/or unstructured objects we automatically partition it into semantically meaningful areas each labeled with a specific object class. We use a novel type of feature which we refer to as a shape filter. Shape filters enable us to capture some or all of shape, texture and appearance context information. A shape filter comprises one or more regions of arbitrary shape, size and 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 we select a sub-set of possible shape filters and incorporate those into a conditional random field model of object classes. That model is then used for object detection and recognition.

Abstract: 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.

Abstract: The present invention is directed to a method for populating a database with a set of images of an anatomical structure. The database is used to perform appearance matching in image pairs of the anatomical structure. A set of image pairs of anatomical structures is received, where each image pair is annotated with a plurality of location-sensitive regions that identify a particular aspect of the anatomical structure. Weak learners are iteratively selected and an image patch is identified. A boosting process is used to identify a strong classifier based on responses to the weak learners applied to the identified image patch for each image pair. The responses comprise a feature response and a location response associated with the image patch. Positive and negative image pairs are generated. The positive and negative image pairs are used to learn a similarity function. The learned similarity function and iteratively selected weak learners are stored in the database.

Abstract: An information processing apparatus includes a feature amount extraction unit extracting a feature amount of each frame of an image, a maximum likelihood state series estimation unit estimating maximum likelihood state series using the feature amount, a highlight label generation unit generating highlight label series with respect to the attention detector learning content, and a learning unit learning the highlight detector that is the state transition probability model using learning label series that is a pair of the maximum likelihood state series obtained from the attention detector learning content and the highlight label series.

Abstract: A crystal lookup table used to define a matching relationship between a signal position of a detected event in a PET scanner and a corresponding detector pixel location is generated using a neural network-based algorithm, and is implemented by a FPGA.

Abstract: A computer-implemented method for non-rigid multi-modal registration, includes providing trained data corresponding to previously registered images captured by different modalities, receiving two or more images captured by the different modalities, and registering the two or more images according to context information determined based on the trained data.

Abstract: An estimation device estimates a hidden state of an estimation subject from an observable state in a manner of a time series. The observable state is observed from the hidden state of the estimation subject under a procedure that has a hierarchical structure, which includes the hidden state of the estimation subject, the observable state, and an intermediate hidden state therebetween. The estimation device includes an estimation subject hidden state predicting means, an intermediate hidden state predicting means based on the state transition structure of the hidden state of the estimation subject, an intermediate hidden state likelihood observing means, an intermediate hidden state estimating means, an estimation subject hidden state likelihood observing means, estimation subject hidden state estimating means, an intermediate hidden state predicting means based on the state transition structure of the intermediate hidden state, and the mixing means.

Abstract: A system and method for detecting metal contraband such as weapons related material in shipping containers where a container is scanned with at least one penetrating beam, preferably a tomographic x-ray beam, and at least one image is formed. The image can be analyzed by a pattern recognizer to find voids representing metal. The voids can be further classified with respect to their 2 or 3-dimensional geometric shapes. Container ID and contents or bill of lading information can be combined along with other parameters such as total container weight to allow a processor to generate a detection probability. The processor can use artificial intelligence methods to classify suspicious containers for manual inspection.

Abstract: A system for automatically classifying an object of a target image is provided. A classification system provides a collection of classified images along with a classification of the dominant object of the image. The classification system attempts to classify the object of a target image based on similarity of the target image to the classified images. To classify a target image, the classification system identifies the classified images of the collection that are most similar to the target image based on similarity between salient points of the target image and the classified images. The classification system selects a classification associated with the classified images that are most similar to the target image as a classification for the object of the target image.

Abstract: A sketch generating system and a method for generating a sketch based on an image are provided. The system includes: a sketch database and a generating subsystem. The sketch database stores local image samples and corresponding local sketch units in different categories. The generating subsystem extracts geometrical features from an input image, retrieves local image units from the input image according to the geometrical features; as to each local image unit retrieved, searches the sketch database for a local sketch unit corresponding to a local image sample having a largest similarity value with the local image unit, and combines all local sketch units found to form one sketch.