Abstract: An apparatus and a method for detecting from an image a particular subject corresponding to multiple views of the subject by dividing a particular subject space into a plurality of subject subspaces and further dividing a subject subspace into subject subspaces representing multiple views; configuring a tree-structured detector wherein the tree structure has a root node that covers all subject subspaces and has a plurality of branches, each branch corresponding to a child node that covers at least one subject subspace; training each node to determine which nodes in the adjacent lower layer the images of the subject in the corresponding nodes should be sent.

Abstract: Multi-scale processing may be used to reduce the memory and computational requirements of optimization algorithms for image labeling, for example, for object segmentation, 3D reconstruction, stereo correspondence, optical flow and other applications. For example, in order to label a large image (or 3D volume) a multi-scale process first solves the problem at a low resolution, obtaining a coarse labeling of an original high resolution problem. This labeling is refined by solving another optimization on a subset of the image elements. In examples, an energy function for a coarse level version of an input image is formed directly from an energy function of the input image. In examples, the subset of image elements may be selected using a measure of confidence in the labeling.

Abstract: An apparatus, method, and medium for detecting a face in an image using classifiers generated by a boost algorithm are provided. The apparatus may include an image dividing unit dividing a black and white image into sub-windows; a sub-window detection unit detecting a sub-window corresponding to a face image in the divided sub-windows using the front part of the cascade of the classifiers and storing the sub-window detected as the face image; and a post-processing unit verifying whether each of sub-windows included in a group among a plurality of groups into which stored sub-windows are grouped is the face image using the number of child detections in each group and the latter part of the cascade of classifiers, integrating the sub-windows in the group according to the result of the verification, and outputting the result of the integration as the face image included in the black and white image.

Abstract: A method and apparatus for encoding and/or decoding depth image-based representation (DIBR) data are provided. The encoding method includes: converting 3-dimensional (3D) volume data into adjustable octree data with predetermined labels given to nodes; by referring to the labels, encoding nodes of the adjustable octree from the root node to leaf nodes by a modified breadth-first search (BFS) method allocating priorities among children nodes; and generating a bitstream with predetermined header information and encoded node data. The decoding method includes: extracting header information containing at least resolution information of an object, from a bitstream and decoding the header information; calculating the number of nodes by using the resolution information of the header, and by a modified BFS method allocating priorities among children nodes, decoding each node of a tree from the root node to leaf nodes; and restoring an adjustable tree by using decoded nodes.

Abstract: The method or system for controlling a printing or copying system for processing single sheets, a first single sheet and at least a second single sheet immediately following the first single sheet is processed with aid of n processing units. A required minimum distance between the first and the second single sheet is determined for each processing unit. A delay distance is determined for the first and at least for each of the n?1 following processing units. A difference between the minimum distance of a processing unit and a sum of the delay distances of the preceding processing units is formed. A smallest possible distance with which the second single sheet is fed to the first processing unit after the first single sheet is determined by a highest value of the determined minimum distance of the first processing unit and the determined differences.

Abstract: Documents from a data stream are clustered by first generating a feature vector for each document. A set of cluster centroids (e.g., feature vectors of their corresponding clusters) are retrieved from a memory based on the feature vector of the document using a locality sensitive hashing function. The centroids may be retrieved by retrieving a set of cluster identifiers from a cluster table, the cluster identifiers each indicative of a respective cluster centroid, and retrieving the cluster centroids corresponding to the retrieved cluster identifiers from a memory. Documents may then be clustered into one or more of the candidate clusters using distance measures from the feature vector of the document to the cluster centroids.

Abstract: A method and system for generating a monotonic classifier are disclosed. A plurality of feature vectors may be mapped to a plurality of ordered classes. Each feature vector may map to a corresponding class. For each class other than the lowest-ordered class, a binary classifier may be trained to distinguish between the class and a directly preceding class in the order. The binary classifiers may be ordered based on the class corresponding to each classifier. The binary classifiers may be ordered in the form of a decision tree, which may be either balanced or unbalanced. A new feature vector pertaining to an individual may then be classified using the binary classifiers.

Abstract: Provided are a method, system, and computer readable medium for training and using classification components on multiple processing units. A plurality of processing units each has a memory including one of a plurality of subsets of a set of data points. At least two of the processing units have different subsets of data points. A plurality of classification components are executed by the processing units. Classification components executing at the processing units are trained, wherein each classification component is trained with the subset of data points in the memory of the processing unit that is executing the classification component. One of the classification components is transferred to an additional processing unit of the processing units to train the transferred classification component using the subset of data points in the memory at the additional processing unit in response to training the classification component with the subset of data points.

Abstract: In an information processing apparatus that processes data using cascade-connected weak classifiers, processing specification information specifying the processing content of each of the weak classifiers is stored. The weak classifiers to be used in processing the data are selected from the weak classifiers by referring to a table in which is specified information for determining the weak classifiers to be used based on a condition for processing the data. The data is then processed by the selected weak classifiers based on the processing specification information that corresponds to those weak classifiers, and an object is extracted from the data using an obtained evaluation value. Through this, a combination of extraction process speed and extraction accuracy can be changed in a flexible manner when extracting a specific object from image data.

Abstract: The distortion of each node is calculated from the distortions of respective elements in the finite element analysis result of a global model of a structure, and a second-order coefficient of a quadratic function representing the displacement at each node of a micro model is calculated from distortions of respective nodes. In addition, a constant term and a first-order coefficient of the quadratic function are calculated from the displacements of the respective nodes of the global model. Then, the displacement at each boundary node of the micro model is calculated using the obtained quadratic function and a finite element analysis of the micro model is performed.

Abstract: A method for creating an appearance model of an object includes receiving an image of the object and creating a hierarchical appearance model of the object from the image of the object. The hierarchical appearance model has a plurality of layers, each layer including one or more nodes. Nodes in each layer contain information of the object with a corresponding level of detail. Nodes in different layers of the hierarchical appearance model correspond to different levels of detail.

Abstract: Provided is an apparatus and method for super-resolution enhancement processing. The apparatus includes a low-resolution image capturing unit, a patch pair generating unit, a patch pair database, and a high-resolution image reconstructing unit. The low-resolution image capturing unit captures a low-resolution image. The patch pair generating unit divides each of a plurality of high-resolution images into patches and generates a plurality of patch pairs composed of medium-frequency patches and high-frequency patches for the patches. The patch pair database stores the plurality of patch pairs generated by the patch pair generating unit.

Abstract: A method is provided for selecting a representative set of training data for training a statistical model in a machine condition monitoring system. The method reduces the time required to choose representative samples from a large data set by using a nearest-neighbor sequential clustering technique in combination with a kd-tree. A distance threshold is used to limit the geometric size the clusters. Each node of the kd-tree is assigned a representative sample from the training data, and similar samples are subsequently discarded.

Abstract: An image indexer for indexing a plurality of images that includes a first data structure for subsequent classification of the one or more images. The first data structure includes characteristics for at least one class. An image classifier classifies one or more individual images found in the plurality of images as classified images according to the first data structure. A second data structure performs subsequent clustering of the plurality of images, wherein the second data structure includes at least two sequential events in a set of known events. The classified images are clustered according to the second data structure, and a representative image is selected from each cluster of classified images.

Abstract: A method of representing a group of images comprises determining the values of one or more dominant colors for the group of images and deriving a dominant color representation expressing the group of images in terms of one or more of said dominant color values.

Abstract: Apparatus for processing image data is provided, comprising first storage means, processing means and manual input means, wherein said first storage means is configured to store said image data and said image data includes a plurality of components defined by a hierarchy of data processing nodes, and said processing means is configured to process each of said nodes in turn. The nodes include standard nodes and switch nodes, each switch node having at least two child nodes one of which is the designated child node of said switch node. The processing means is configured to process a standard node only when each of its child nodes has been processed, and process a switch node when only its designated child node has been processed. Additionally, said processing means may include second storage means and stores, in either said first or said second storage means, information recording the states of said switch nodes that defines a version of said image data.

Abstract: A combination of inter-frame and intra-frame and sequential classification methods provide for robust and flexible object classification. Intra-frame classification determines the object type based on features extracted from a single image. Inter-frame classification determines object types based on the features extracted from several sequential frames. A sequential classification combines the results of the inter- and intra-frame classifications over a sequence of images.

Abstract: Image recognition is utilized to facilitate in scoring parse trees for two-dimensional recognition tasks. Trees and subtrees are rendered as images and then utilized to determine parsing scores. Other instances of the subject invention can incorporate additional features such as stroke curvature and/or nearby white space as rendered images as well. Geometric constraints can also be employed to increase performance of a parsing process, substantially improving parsing speed, some even resolvable in polynomial time. Additional performance enhancements can be achieved in yet other instances of the subject invention by employing constellations of integral images and/or integral images of document features.

Abstract: A method provided for the formation of a selective rendering of body structures of an object to be examined from a primary image data set. The method provides a simple and fast selection or deselection of given body structures. The method comprises the following steps: forming at least one pixel group (BG1-BG16) which comprises pixels from the primary data set which are in conformity with predetermined filter criteria, forming at least one pixel list by selection and/or deselection of at least one pixel group in conformity with predetermined criteria, marking the pixels of the pixel groups of the at least one pixel list, forming a filtered secondary image data set which includes the marked pixels, and forming the rendering from the secondary image data set, the marked pixels being separately rendered in highlighted or suppressed form.

Abstract: Processing for improving the contrast of an image of interest is executed by an image processing circuit in an X-ray photography apparatus. The image is decomposed into coefficient values of a plurality of frequency bands by a frequency-component decomposition circuit to obtain coefficients of each and every frequency band. A coefficient conversion circuit converts coefficient values with regard to at least one frequency band of the plurality of frequency bands in such a manner that a frequency distribution of coefficient values of every frequency band will become a prescribed frequency distribution. A restoration circuit generates an image using the coefficient values obtained by the conversion performed by the coefficient conversion circuit.

Abstract: Information associated with data which was previously transmitted to a client terminal (201 or 202) is recorded as a history. When a transmission request of data of logical units in tiles required to obtain a desired image is received from the client terminal (201), the type of progression order used in the client terminal (201) is discriminated with reference to the history. The transmission order of the data of the logical units in the tiles to be transmitted to the client terminal (201) is determined in accordance with the discrimination order, and the data of the logical units in the tiles are transmitted to the client terminal (201) in accordance with the determined transmission order.

Abstract: A system (400) for generating training images for a pattern recognition classifier (54) from three-dimensional image data representing an output class comprises a model constructor (404) that constructs an image model in a standard coordinate domain from the three-dimensional image data. A model transformer (406) manipulates the image model in the standard coordinate domain to produce at least one manipulated image model. An image restorer (408) maps the at least one manipulated image model from the standard coordinate domain to a coordinate domain associated with an image source utilized during the run-time operation of the pattern recognition classifier.

Abstract: A system and method for tree-model visualization for detecting an abnormality in an anatomical tree structure are provided. The method comprises: fitting a tree-model to an anatomical tree structure; converting branches of the tree-model into first two-dimensional branch images; and arranging the first two-dimensional branch images in a hierarchical order to form a second two-dimensional image.

Abstract: A method of processing digitized picture data includes providing an hierarchical structure graph which is applied to the picture data in that, starting with the lowermost level, at least one node is processed. The provided hierarchical structure graph includes nodes, a specified number of levels, with at least one node located in each level, and edges connecting pairs of predetermined nodes of different levels and defining, for each pair of nodes, a lower father node and an upper son node. Processing of a node includes matching of its at least one picture matching model to the picture data by variation of the model parameters, determination of a matching quantity for each parameter variation, and determination of an assessment for each parameter variation. The assessment found for each parameter variation is applied as a criterion for the processing of a son node of the processed node.

Abstract: Data values, such as residual values based on a predictive value, are compressed based on a number of significant digits used to represent the data value that is termed a magnitude level. Data values are grouped, and a highest magnitude level of the magnitude levels associated with each of the data values is associated with each group. The magnitude level is expressible in fewer digits than a value of that magnitude. Further, selecting one magnitude level for each group reduces a number of magnitude levels stored, further reducing the size of the resulting file. Choosing a magnitude level associated with a group of related data values in most cases results in the highest magnitude level being close to the magnitude level for each of the data values it represents, resulting in few bits being wasted in coding the data values.

Abstract: A method of identifying at least one particular person of interest in a collection of digital images taken over time, includes providing the collection of digital images with each digital image containing one or more persons; storing in a digital database an appearance model having a set of features associated with the particular person of interest and a prior probability of the person of interest appearing in images by a particular photographer; and comparing the appearance model with features extracted from the digital images in the collection of digital images and using the prior probability and the appearance model to determine if the particular person of interest is in one or more digital images in the collection of digital images.

Abstract: There is provided a signal generator which includes a main memory which stores at least time data indicating timing when a state change of a timing pulse occurs and space data indicating a state of the timing pulse corresponding to the time data; a counter which counts a clock to be a reference; a first memory which stores predetermined time data in time data stored in the main memory; a comparator which compares an output of the counter with the predetermined time data stored in the first memory and to output a result; a second memory which stores predetermined space data in space data stored in the main memory; and output signal control means which controls the timing pulse to a state of the timing pulse indicated by the space data stored in the second memory on the basis of an output of the comparator.

Abstract: Methods and apparatuses for blending two images using vector table look up operations. In one aspect of the invention, a method to blend two images includes: loading a vector of keys into a vector register; converting the vector of keys into a first vector of blending factors for the first image and a second vector of blending factors for the second image using a plurality of look up tables; and computing an image attribute for the blended image using the blending factors.

Abstract: A classifier, method and computer program product for categorizing events within a data stream includes a correlation engine having at least one set of rules, the rules including event selectors with logical expressions corresponding to a plurality of attributes for the events within the data stream; the engine also including a decision tree built by the engine for providing an index of the rules, wherein the engine references the decision tree to access selected rules corresponding to the events and applies the selected rules to correlate the events.

Abstract: A method for classifying features in a digital medical image includes providing a plurality of feature points in an N-dimensional space, wherein each feature point is a member of one of two sets, determining a classifying plane that separates feature points in a first of the two sets from feature points in a second of the two sets, transforming the classifying plane wherein a normal vector to said transformed classifying plane has positive coefficients and a feature domain for one or more feature points of one set is a unit hypercube in a transformed space having n axes, obtaining an upper bound along each of the n-axes of the unit hypercube, inversely transforming said upper bound to obtain a new rule containing one or more feature points of said one set, and removing the feature points contained by said new rule from said one set.

Abstract: A vehicle restraint system has a vision-based occupant classification system for control of airbag deployment during a crash scenario. The classification system utilizes two imaging sensors which together create a stream of paired images received and stored by an occupant classification controller. A computer program product of the controller utilizes the paired images to extract disparity/range features and stereo-vision differential edge density features. Moreover, the controller extracts wavelet features from one of the two paired images. All three features or maps are classified amongst preferably seven classifications by algorithms of the computer program product producing class confidence data fed to a sensor fusion engine of the controller for processing and output of an airbag control signal input into a restraint controller of the vehicle restraint system.

Abstract: The present invention includes a system and method for automatically extracting the demographic information from images. The system detects the face in an image, locates different components, extracts component features, and then classifies the components to identify the age, gender, or ethnicity of the person(s) in the image. Using components for demographic classification gives better results as compared to currently known techniques. Moreover, the described system and technique can be used to extract demographic information in more robust manner than currently known methods, in environments where high degree of variability in size, shape, color, texture, pose, and occlusion exists. This invention also performs classifier fusion using Data Level fusion and Multi-level classification for fusing results of various component demographic classifiers.

Abstract: Systems and methods are provided for constructing a classification architecture for a pattern recognition classifier that determines an associated occupant class of an occupant within a vehicle. A plurality of occupant classes are defined, with at least one of the occupant classes having an associated set of subclasses (202). A hierarchy of subclass partitions is determined based on a degree of similarity between subclasses associated with a given occupant class (210). Each successive partition in the hierarchy is an immediate partition of the preceding partition. The following is then performed iteratively until a termination event occurs. A subclass partition is selected according to the determined hierarchy (216/230). The fitness of a classifier model having an architecture derived from the selected subclass partition is evaluated (220).

Abstract: A category classification apparatus includes: an overall classifier that classifies a category to which an image belongs, based on an overall characteristic amount that is obtained from image data, the overall characteristic amount indicating an overall characteristic of the image represented by the image data; and a partial classifier that classifies a category to which the image belongs, based on partial characteristic amounts that are obtained from partial image data included in the image data, the partial characteristic amounts indicating characteristics of portions of the image.

Abstract: A method and apparatus for the detection of local image structures represented as clusters in a joint-spatial range domain where the method comprises receiving an input image made having one or more clusters in a joint-spatial range domain, and each of the one or more clusters having a corresponding mode. Receiving a set of analysis matrices and selecting through each one of the analysis matrices. Using the selected analysis matrix to partition the input image into the one or more clusters and their corresponding modes, and computing a mean, ?, and a local covariance matrix ? for each of the corresponding modes of each of the one or more clusters. Selecting at least one of the one or more clusters, where each selected cluster has a stable mean and stable covariance matrix across the set of analysis matrices, whereby each of the selected clusters is indicative of a local image structure.

Abstract: A method of representing a data distribution derived from an object or image by processing signals corresponding to the object or image comprising deriving an approximate representation of the data distribution and analysing the errors of the data elements when expressed in terms of the approximate representation.

Abstract: The present invention relates to a method and system for generating aesthetic characters, and to a business model for commercially developing the method and system utilizing the Internet or the like. The present invention makes it possible to synthesize a variety of image patterns with arbitrary characters by adding stroke fonts. According to a first aspect, the present invention provides a method and system for generating aesthetic characters, in which an arbitrary image pattern is selected from a pattern database, and data on arbitrary characters constituting single lines are superposed on the selected image pattern, whereby the arbitrary image patterns are processed into various characters and generating aesthetic image patterns. Thus, it is possible to provide character bodies drawn with excellent aesthetic properties that are uniquely designed for attracting public attention, such as in the television broadcasts, Internet sites, headings of journals or newspapers, advertisements, and signboards.

Abstract: An object image detection device is disclosed that is able to rapidly detect an object image from an input image without a great deal of computation. The object image detection device includes an object image classification unit for determining whether the object images are included in an image having a given orientation, an image orientation detection unit for detecting orientation of the input image, an image rotation unit for rotating the object image classification unit according to the detected orientation of the input image, and a detection unit for detecting the object images from the input image by using the rotated object image classification unit.

Abstract: Automatic detection and tracking of multiple individuals includes receiving a frame of video and/or audio content and identifying a candidate area for a new face region in the frame. One or more hierarchical verification levels are used to verify whether a human face is in the candidate area, and an indication made that the candidate area includes a face if the one or more hierarchical verification levels verify that a human face is in the candidate area. A plurality of audio and/or video cues are used to track each verified face in the video content from frame to frame.

Abstract: Performing Boolean operations among two regions; regions and results represented as vector tuples. Establish indexing cells about regions and classify by type of interaction between regions, e.g., (boundary, boundary). For each (boundary, boundary) cell, define pseudo-points at each boundary entrance/exit. Categorize each (boundary, boundary) cell on relationship of its pseudo-points. Identify starting points along boundaries based on: categorization, operation, and interior convention. Accumulate results cycling from a starting point, along region boundary. Upon encountering each intersection, proceed along the other region boundary. Upon encountering a cell edge, proceed along the cell edge in the direction consistent with the interior convention. Where a starting point or intersection between region boundaries remains untraversed, accumulate results in tracing a similar cycle along untraced boundaries. Discard duplicate/cell-edge only tuples.

Abstract: A method and apparatus for determining a closest match of N input patterns relative to R reference patterns using K processing units. Each of a set of input patterns are loaded into the K processing units. One of the Reference patterns is sequentially loaded into each of the processing units and a distance defining the similarity between the reference pattern and each of the input patterns is calculated. A present calculated distance replaces its corresponding stored present minimum distance if it is has a smaller value. After the R reference patterns have been processed the minimum distance and its corresponding identification for all N input patterns is determined without merging outputs. The minimum distances and the identifications may be read either in parallel or serially. The apparatus is easily scalable by adding processors. The number of reference patterns may be easily increased without altering system configuration.

Abstract: An improved method and mechanism for recognizing chirographs (handwritten characters) input into a computer system. A primary recognizer is provided for converting chirographs to code points, and secondary recognizers such as binary CART trees are developed and trained to differentiate chirographs which produce certain code points at the primary recognizer. Each such secondary recognizer is associated with each selected code point. When a chirograph is received, the chirograph is provided to the primary recognizer whereby a code point corresponding thereto is received. If the code point corresponds to one of the secondary recognizers, the chirograph is passed to the secondary recognizer, and a code point is returned from the secondary recognizer. If not, the code point provided by the primary recognizer is returned.

Abstract: Methods and apparatus, including computer program products, implementing techniques for testing user interfaces of software application. The techniques include interacting with a user to identify one or more straight lines in a first rendering of a user interface for a software application, the straight lines being diagnostic of the correct layout of the first rendering of the user interface, and using the one or more straight lines as reference lines to determine if a second rendering of the user interface has a correct layout.

Abstract: A method is provided for coding a source mesh (M) representing a three-dimensional object or a scene, using a simple mesh (M0) having a limited number of facets, each defined by vertices and edges, and coefficients in a base of wavelets, corresponding to local modifications of the simple mesh (M0), and making it possible, at each instant, to determine a function of parameterization of the mesh. The method includes a step to eliminate the effect of at least one wavelet coefficient (d; d1, . . . , dn) verifying at least one criterion of non-essentiality, so as to give a partial mesh (M?), corresponding to a partial representation of the source mesh (M).

Abstract: The present invention makes it possible to effectively encode/decode position information of leaves or nodes in tree-structured information by using a less amount of bits. The coding device according to the present invention includes a branching layer discriminating portion for identifying a hierarchical layer in which a leaf or node and a preceding leaf or node are separately branched and a position information coding portion for outputting a code corresponding to the identified hierarchical layer as a position information code for the leaf or node. Namely, merely information indicating a hierarchical layer in which a current leaf or node and a preceding leaf or node are separately branched is encoded as the position information of the leaf or node, thus increasing the efficiency of encoding leaf or node positions in the tree-structured information.

Abstract: The present invention relates generally to the classification of items into categories, and more generally, to the automatic selection of different classifiers at different places within a hierarchy of categories. An exemplary hierarchical categorization method uses a hybrid of different classification technologies, with training-data based machine-learning classifiers preferably being used in those portions of the hierarchy above a dynamically defined boundary in which adequate training data is available, and with a-priori classification rules not requiring any such training-data being used below that boundary, thereby providing a novel hybrid categorization technology that is capable of leveraging the strengths of its components.

Abstract: Processing for improving the contrast of an image of interest is executed by an image processing circuit in an X-ray photography apparatus. The image is decomposed into coefficient values of a plurality of frequency bands by a frequency-component decomposition circuit to obtain coefficients of each and every frequency band. A coefficient conversion circuit converts coefficient values with regard to at least one frequency band of the plurality of frequency bands in such a manner that a frequency distribution of coefficient values of every frequency band will become a prescribed frequency distribution. A restoration circuit generates an image using the coefficient values obtained by the conversion performed by the coefficient conversion circuit.

Abstract: The present invention includes a method and system for automatically extracting the multi-class age category information of a person from digital images. The system detects the face of the person(s) in an image, extracts features from the face(s), and then classifies into one of the multiple age categories. Using appearance information from the entire face gives better results as compared to currently known techniques. Moreover, the described technique can be used to extract age category information in more robust manner than currently known methods, in environments with a high degree of variability in illumination, pose and presence of occlusion. Besides use as an automated data collection system wherein given the necessary facial information as the data, the age category of the person is determined automatically, the method could also be used for targeting certain age-groups in advertisements, surveillance, human computer interaction, security enhancements and immersive computer games.

Abstract: A system and method for classifying facial image data, the method comprising the steps of: training a classifier device for recognizing one or more facial images and obtaining corresponding learned models the facial images used for training; inputting a vector including data representing a portion of an unknown facial image to be recognized into the classifier; classifying the portion of the unknown facial image according to a classification method; repeating inputting and classifying steps using a different portion of the unknown facial image at each iteration; and, identifying a single class result from the different portions input to the classifier.

Abstract: Document processing and understanding can be important for a variety of applications such as office automation, creation of electronic manuals, online documentation and annotation, etc. The first step towards this process can involve the decomposition of a page into separate units describing the text, graphics, pictures etc. Towards that end, a method of page decomposition comprises segmenting an image into a plurality of blocks according to an amount of information contained in each block, classifying each block according to a corresponding frequency spectrum, and refining a classification of each block.