Abstract: A method for selecting important digital images in a collection of digital images, comprising: analyzing the digital images in the collection of digital images to identify one or more sets of similar digital images; identifying one or more sets of similar digital images having the largest number of similar digital images; selecting one or more digital images from the identified largest sets of similar digital images to be important digital images; and storing an indication of the selected important digital image in a processor accessible memory.

Abstract: There is provided an image processing apparatus that analyzes a pathology image obtained by imaging an analysis object including a setting device for calculating an amount of a luminance difference characteristic as an amount of characteristic of each pixel of a pathology image, and for setting a cellular tissue region in the pathology image, based on the amount of the luminance difference characteristic, using a cellular tissue detector that has studied in advance. The cellular tissue detector may have studied in advance by statistical learning that uses an image corresponding to a cellular tissue.

Abstract: A method of quantifying social affinity from multiple images includes identifying each image showing both a first person and a second person and determining a weighted affinity value between the first person and the second person. The weighted affinity value is determined based on a total number of persons appearing in each identified image, a physical distance between the first person and the second person represented in each identified image, and a total number of identified images.

Abstract: An image analyzer (120) aggregates image samples (140) into a cluster (170) based on the image samples (140) being classified from a subset of image metrics applied to a reference sample (130). The image analyzer (120) generates an image quality output (150) by analyzing a distance (180) from the cluster (180) relative to another cluster.

Abstract: A method for searching in a large-scale database of images for images that are visually similar to query images is provided using a processor to perform the following steps: computing visual features for the database images; constructing an Image Anchor Graph that contains a plurality of anchors corresponding to a subset of images in the database, according to the computed visual features; computing visual features for the query images; and retrieving database images similar to one or more user query images using a ranking algorithm that uses the anchors in the Image Anchor Graph according to the computed visual features of both the query images and the database images on the Image Anchor Graph.

Abstract: A computing device segments an image into a plurality of segments, wherein each segment of the plurality of segments comprises a set of pixels that share visual characteristics. The computing device then determines expected contexts for the segments, wherein an expected context for a segment comprises at least one of additional segments or features expected to occur in the image together with the segment. The computing device then identifies a probable object based on the expected contexts.

Abstract: A method and apparatus for providing targeted advertisements is provided herein. In particular, targeted advertisements are provided to users based on face clustering for time-varying video. During operation video is continuously obtained of users of the system. Users' faces are detected and measured. Measurements of users' faces are then clustered. Once the clusters are available, advertisements are targeted at clusters rather than individual users.

Abstract: Technologies are generally described for an general mood adding scheme in a cloud-based game system. In some examples, a method performed under control of a computing device may include receiving from a raw database at least one set of facial expression data, each of the at least one set of facial expression data being accompanied by time information and location information, clustering a geographic area to form at least one cluster based at least in part on the at least one set of facial expression data, and storing the at least one cluster in a map database corresponding to the area.

Abstract: 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.

Abstract: Image search techniques are described. In one or more implementations, images in a search result are ordered based at least in part on similarity of the images, one to another. The search result having the ordered images is provided in response to a search request.

Abstract: Briefly, in accordance with one aspect, a method of binarizing an image is provided. The method includes partitioning the image into a plurality of image segments, each image segment having a plurality of image pixels and partitioning each of the image segments into subsegments, each image subsegment having a plurality of image pixels. The method also includes estimating a grey membership parameter for each image pixel for each of the plurality of image segments and subsegments, combining the grey membership parameter for each of the plurality of image pixels from each of the plurality of image segments and subsegments to estimate a net grey membership parameter for each image pixel and assigning black or white color to each of the plurality of image pixels based on the estimated net grey membership parameter of the respective pixel.

Abstract: Illumination information indicating the center of illumination and lightness change direction is acquired from an image, and the image is divided into regions for respective colors based on the acquired illumination information and color information of respective pixels included in the image. Especially, thresholds in clustering are set step by step in accordance with the lightness change direction from the center of the illumination information, and the image is clustered to the regions for respective colors according to the thresholds which are set step by step.

Abstract: A computer-implemented system and method for organizing cluster groups within a display is provided. Cluster groups each having one or more spines of clusters are obtained. Each cluster includes at least one document. At least one of the cluster groups is placed around a ring centrally defined in a display. Circle sectors are defined around the ring and an initial target angle is identified within each of the sectors. The at least one cluster group is positioned within one of the circle sectors at the initial target angle. A further one of the cluster groups is positioned within a different circle sector up to one of a maximum and minimum angle relative to the initial target angle for that sector.

Abstract: The invention relates to a method for processing video signals from a video sensor, in order to extract 3d shape information about objects represented in the video signals, the method comprising the following steps: providing a memory in which objects are stored in a 3d shape space, the shape space being an abstract feature space encoding the objects' 3d shape properties, and mapping a 2d video signal representation of an object in the shape space, the coordinates of the object in the shape space indicating the object's 3d shape.

Abstract: A method of converting a raster scan image to a vector-based image including (a) reading a scanline of a raster scan image containing n pixels per scanline; (b) classifying n+1 groupings of four adjacent pixels within the scanline using only the four adjacent pixel values; (c) assigning to each grouping (x,y), for all groupings in the scanline, plural vectors based on (1) the classification of the grouping (x,y) and (2) vectors assigned to a grouping (x?1,y) and vectors assigned to a grouping (x, y?1); (d) adding a point (x,y) to at least one vector assigned to the grouping (x,y) when the classification of grouping (x,y) indicates that the point (x,y) corresponds to a corner; (e) determining whether the points added to any vector form a closed loop; (f) writing out the points added to any vector that form a closed loop; and (g) repeating steps (a)-(f) for all of the scanlines of the raster scan image.

Abstract: A system, method, and computer program product for detecting anomalies in an image. In an example embodiment the method includes partitioning each image of a set of images into a plurality of image local units. The method further includes clustering all local units in the image set into clusters, and consequently assigning a class label to each local unit based on the clustering results. The local units with identical class labels having at least one substantially related image feature. Further, the method includes assigning a weight to each of the local units based on a variation of the class labels across all images in a set of images. The method further includes performing a clustering over all images in the set by using a distance metric that takes the learned weight of each local unit into account, then determining the images that belong to minorities of the clusters as anomalies.

Abstract: Methods and systems for selecting a representative image of an entity are disclosed. The representative image can be, but is not limited to, that of a person, a corporate logo, other symbol associated with a person or corporation. According to one embodiment, a computer-implemented method for selecting a representative image of an entity is disclosed. The method includes: accessing a collection of images of the entity; clustering, based on similarity of one or more similarity features, images from the collection to form a plurality of similarity clusters; and selecting the representative image from one of said similarity clusters. Further, based on cluster size of said similarity clusters popular clusters can be determined, and the selection of the representative image can be from the popular clusters.

Abstract: A document recognition system and method, where images are represented as a collection of primitive features whose spatial relations are represented as a graph. Useful subsets of all the possible subgraphs representing different portions of images are represented over a corpus of many images. The data structure is a lattice of subgraphs, and algorithms are provided means to build and use the graph lattice efficiently and effectively.

Abstract: There is provided an information processing device, including a cluster reconfiguration determining unit that, in a case where new data having positional information in a feature amount space is added to a data group in which each data has positional information in the feature amount space and is classified into a cluster based on a distance in the feature amount space, decides that a candidate cluster is to be reconfigured when the new data is classified into the cluster if there are a plurality of candidate clusters into which the new data is classifiable based on the distance in the feature amount space among the clusters.

Abstract: A method and apparatus for segmenting an image are provided. The method may include the steps of clustering pixels from one of a plurality of images into one or more segments, determining one or more unstable segments changing by more than a predetermined threshold from a prior of the plurality of images, determining one or more segments transitioning from an unstable to a stable segment, determining depth for one or more of the one or more segments that have changed by more than the predetermined threshold, determining depth for one or more of the one or more transitioning segments, and combining the determined depth for the one or more unstable segments and the one or more transitioning segments with a predetermined depth of all segments changing less than the predetermined threshold from the prior of the plurality of images.

Abstract: A method for organizing an event timeline for a digital image collection, includes using a processor for detecting events in the digital image collection and each event's associated timespan; determining the detected events that are significant in the digital image collection; and organizing the event timeline so that the event timeline shows the significant events and a clustered representation of the other events, made available to the user at different time granularities. The organized event timeline is also used for selecting images for generating output.

Abstract: A shape of an object is represented by a set of points inside and outside the shape. A decision function is learned from the set of points an object. Feature points in the set of points are selected using the decision function, or a gradient of the decision function, and then a local descriptor is determined for each feature point.

Abstract: Methods and apparatuses for constructing a multi-level solver, comprising decomposing a graph into a plurality of pieces, wherein each of the pieces has a plurality of edges and a plurality of interface nodes, and wherein the interface nodes in the graph are fewer in number than the edges in the graph; producing a local preconditioner for each of the pieces; and aggregating the local preconditioners to form a global preconditioner.

Abstract: A method for feature matching in image recognition is provided. First, image scaling may be based on a feature distribution across scale spaces for an image to estimate image size/resolution, where peak(s) in the keypoint distribution at different scales is used to track a dominant image scale and roughly track object sizes. Second, instead of using all detected features in an image for feature matching, keypoints may be pruned based on cluster density and/or the scale level in which the keypoints are detected. Keypoints falling within high-density clusters may be preferred over features falling within lower density clusters for purposes of feature matching. Third, inlier-to-outlier keypoint ratios are increased by spatially constraining keypoints into clusters in order to reduce or avoid geometric consistency checking for the image.

Abstract: A method, system and computer program for segmenting an image are provided. The method may include the steps of acquiring a plurality of images, and clustering pixels from one of the plurality of images into one or more segments, each segment being defined in a state of either stable, mesostable or unstable. The clustering may include the steps of determining residual pixels changing by more than a predetermined threshold, grouping the determined residual pixels to form segments selected from the group of unstable and mesostable segments, determining depth for all unstable and mesostable segments, transitioning unstable segments to corresponding mesostable segments if predetermined stability criteria are met, transitioning mesostable segments to corresponding stable segments upon a determination that the mesostable segments have remained stable for a predetermined number of frames, and merging stable and mesostable segments.

Abstract: A method of detecting recurring events in a digital image collection taken over a pre-determined period of time is disclosed. The method uses a processor for analyzing the digital image collection to produce a two-dimensional representation of the distribution of image capture activity over time and detecting recurring events by identifying spatial clusters in the two-dimensional representation.

Abstract: An apparatus for providing multi-feature based sampling for relevance feedback may include a user interface element. The user interface element may be configured to receive a query image, provide a first set of images determined to be relevant with respect to the query image, and provide a second set of images in which each image of the second set of images comprises a first relevance determination with respect to a first feature of an image and a second relevance determination for at least a second feature of the image. Each image of at least the second set of images may include an associated user interface item to enable a user to provide relevance feedback with respect to each image of the second set of images.

Abstract: A method and system for automatic detection and volumetric quantification of bone lesions in 3D medical images, such as 3D computed tomography (CT) volumes, is disclosed. Regions of interest corresponding to bone regions are detected in a 3D medical image. Bone lesions are detected in the regions of interest using a cascade of trained detectors. The cascade of trained detectors automatically detects lesion centers and then estimates lesion size in all three spatial axes. A hierarchical multi-scale approach is used to detect bone lesions using a cascade of detectors on multiple levels of a resolution pyramid of the 3D medical image.

Abstract: Image classification methods and systems are provided. First, an image is obtained using a computer. The image is then processed using the computer to obtain image information. The image information includes one or any combination of an average color difference between at least one average channel value of pixels in at least one boundary region of the image and a predefined standard value, a gradient variation magnitude difference between at least two regions of the image, and a percentage of the edges of the image to the image. The image is classified using the computer according to the image information.

Abstract: Disclosed is an image processing apparatus, method and computer-readable medium of a robot which efficiently manage a moving image acquired by the robot and reinforce security to prevent image leakage. In order to restore an object region within an original image photographed by the robot, a low-resolution image is generated using object password information and the original image to generate image information, and the image information is transmitted to a server over a network. The server detects the object password information and the low-resolution information from the image information and restores a high-resolution object image using the object password information and the low-resolution image.

Abstract: The present disclosure relates to an assistance device for image recognition that comprises a memory storing sets of image descriptors, respectively associated with an image area and including first vector data, second angle data, and third scale data. A first operator receives a designation of two descriptors and establishes a Boolean representing a check of a similarity criterion of the descriptor vectors from a comparison among first data. A second operator receives a designation of two descriptors and establishes a rotation angle parameter from the second data. A third operator receives a designation of two descriptors and establishes a scale factor parameter from the third data.

Abstract: A method for recognizing markers printed on a learning material, includes sampling an image of the learning material; grouping the sampled image of pixels into a first image group and a second image group based on a threshold; and calculating medians of the first image group and the second image group to update the threshold with a first average value of the calculated medians. Further, the method for recognizing markers printed on the learning material includes repeating the above until a difference between a previous threshold and an updated threshold is equal to or smaller than a reference value; binarizing an image captured by a camera based on the updated threshold; and detecting the markers based on the binary image.

Abstract: An information processing device includes: an extracting unit configured to extract a face from an image; a grouping unit configured to classify the image extracted by the extracting unit into a group; an adding unit configured to add to a group generated by the grouping unit a tag indicating relationship between persons within the group; a closeness calculating unit configured to calculate the closeness of the person of the face from distance between groups generated by the grouping unit; and a person correlation data generating unit configured to generate person correlation data including at least the tag and the closeness.

Abstract: A system for automatically extracting or isolating structures or areas of interest (e.g., built-up structures such as buildings, houses, shelters, tents; agricultural areas; etc.) from HR/VHR overhead imagery data by way of making as little as a single pass through a hierarchical data structure of input image components (where pixels are grouped into components based on any appropriate definition or measure of dissimilarity between adjacent pixels of the input image) to identify candidate components (e.g., possible structures of interest) free of necessarily having to re-iterate the same operator configured with different threshold parameters for a plurality of values.

Abstract: This disclosure describes various exemplary systems, computer program products, and methods for feature distance metric learning with feature decomposition (DMLFD). The disclosure describes decomposing a high-dimensional feature space into one or more low-dimensional feature spaces according to minimum dependence. Furthermore, the disclosure describes how the sub-metrics are constructed and combined to form a global metric.

Abstract: A person cluster selection part selects one or more person clusters as representative person clusters in order of largest number of facial images included therein. A unit selection part selects, from each representative person cluster, (i) a unit having the highest likelihood as a first representative unit, and (ii) units in order of lowest likelihood as a second representative unit onward. A representative facial image selection part selects, from each representative unit, (i) a facial image having the highest likelihood as a first representative facial image, and (ii) facial images in order of lowest likelihood as a second representative facial image onward.

Abstract: A classification result is stored in advance. At the time of identification, classification is performed at a high speed based on the stored classification result. In pre-classification, classification is performed divisionally in two stages. In the first classification, a plurality of representative vectors are generated from the classification result using a fast executable method. In the second classification, the created representative vectors are classified, and the result is stored in a lookup table. When an identification target image is input, the classification result can be reflected at a high speed by using the lookup table.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for graphical overlay object translation. One of the methods includes receiving a first plurality of first feature points, each first feature point being associated with a plurality of feature descriptors and a respective location on the first digital image. The method includes receiving a second plurality of second feature points, each second feature point being associated with a plurality of feature descriptors and a respective location on the second digital image. The method includes determining the translation offset for the second digital image from the first digital image by matching locations of the first feature points with locations of the second feature points without regard to the feature descriptors.

Abstract: Systems, methods and computer program products on storage devices for shape clustering and applications in processing various documents, including an output of an optical character recognition (OCR) process. The output of an OCR process is classified into a plurality of clusters of clip images and a representative image for each cluster is generated to identify clusters whose clip images were incorrectly assigned character codes by the OCR process.

Abstract: Facilitating access to content in a content-aware mesh may include a coordinating node storing information associated with each image of a plurality of images. The images may be distributed among at least two nodes of the mesh. The coordinating node may provide access to each image to a node of the mesh. Providing access may be performed without transmitting all of the images to the node. The node may be configured to display an image of the plurality of images via a uniform interface without indication as to which of the at least two nodes the image is stored on.

Abstract: 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.

Abstract: Clustering algorithms such as k-means clustering algorithm are used in applications that process entities with spatial and/or temporal characteristics, for example, media objects representing audio, video, or graphical data. Feature vectors representing characteristics of the entities are partitioned using clustering methods that produce results sensitive to an initial set of cluster seeds. The set of initial cluster seeds is generated using principal component analysis of either the complete feature vector set or a subset thereof. The feature vector set is divided into a desired number of initial clusters and a seed determined from each initial cluster.

Abstract: In one aspect, a method to reduce false alarms in identifying whether a candidate image is from an object class includes projecting the candidate image onto an object class subspace and projecting the candidate image onto a non-object class subspace. The method also includes determining whether the candidate image is from the object class using a Bayesian decision function based on the projections on the object class subspace and the non-object class subspace. In another aspect, a method to reduce false alarms in identifying whether a candidate mine image is from a mine class includes projecting the candidate mine image onto a mine subspace and projecting the candidate mine image onto a non-mine subspace. The method also includes determining whether the candidate mine image represents a mine using a Bayesian decision function based on the projections on the mine class subspace and the non-mine class subspace.

Abstract: A method for creating an image database comprising an extraction step of extracting reference feature vectors from a reference image which should be compared with a retrieval query image for object recognition, the reference feature vectors corresponding to local features at different positions of the reference image and representing the position and characteristics of each of the local features as a vector position, vector length, and a vector direction, a clustering step of creating a plurality of clusters consisting of different reference feature vectors in such a manner that each reference vector belongs to any of the plurality of clusters, a selection step of selecting the representative vector of the clusters from among the reference feature vectors of each of the clusters, and a step of associating the representative vector with the reference image and registering the representative vector associated therewith in the image database for the object recognition, wherein, in the clustering step, each of th

Abstract: Systems and methods for clustering a plurality of feature vectors. A hierarchical clustering algorithm is performed on the plurality of feature vectors to provide a plurality of clusters and a cluster similarity measure for each cluster representing the quality of the cluster. Each cluster of the plurality of clusters with a cluster similarity measure meeting a threshold value is accepted. A clustering algorithm is performed on each cluster that fails to meet the threshold value to provide a set of subclusters each having an associated cluster similarity measure. Each subcluster having a cluster similarity measure meeting the threshold value is accepted.

Abstract: Embodiments include a scene classification system and method. In one embodiment, a method includes forming a first plurality of image features from an input image, processing the first plurality of image features in the first scene classifier.

Abstract: A system and method for generating groups of cluster spines for display are provided. One or more concepts for each cluster in a set of clusters are generated. Spines are formed from at least a portion of the clusters based on the concepts. At least one spine unique from all other spines is placed. One or more spine groups are generated by positioning at least one unplaced spine in relation to one of the placed unique spines. The spine groups are displayed.

Abstract: Automatic stacking based on time proximity and visual similarity is described, including a method, comprising analyzing a time proximity of a plurality of electronic images, performing a visual similarity analysis on the plurality of electronic images, and stacking the plurality of electronic images based on a result of the time proximity analysis and the visual similarity analysis.

Abstract: A system and method for information retrieval comprises a computing device with a client and a progressive feature server coupled by a network. The client includes a feature extraction module, a progressive sending module, a sampling module and a feedback receiver. The progressive sending module divides the features into groups and sends them to the server. The progressive sending module is also responsive to commands from the feedback receiver to terminate the transmission of additional features if a match is found. The progressive feature server is coupled for communication with the client and includes a progressive receiving module, a feedback generator, an image recognizer and a result delivery module. The image recognizer progressively processes the groups of extracted features received from the client and provides results and hints for improved recognition.

Abstract: A method of detecting a target in an image. The method includes receiving an image; generating a plurality of scaled-down images based on the received image; generating integral column images of each of the plurality of scaled images by calculating integral values of pixels column by column; selecting and classifying a plurality of windows of the integral column images according to a feature arithmetic operation based on a recursive column calculation; and detecting the target on the basis of the classification results for the plurality of windows.