Abstract: A method for extracting an epitome from an image Y is described. The method comprises: searching, for each block, a list of patches in the image Y with similar content; determining, for each patch of said lists, the set of blocks able to be represented by the patch; constructing, from the sets of blocks, epitome charts comprising, for each epitome chart, initializing said epitome chart with one patch and extending said epitome chart with at least one candidate region contained in a patch overlapping said epitome chart; said method being characterized in that said candidate region is selected by taking into account the blocks represented by said patch overlapping said epitome chart and by any other patch contained in said epitome chart enlarged by said selected candidate region.

Abstract: A computer-implemented method for object recognition using a recursive cortical network comprising receiving an input image at an input module, applying a trained recursive cortical network (RCN) to the image using an inference module to activate child features of the RCN, selecting pools of the RCN containing the activated child features, propagating the selection of the pools to identify probabilities of one or more high-level features matching one or more objects in the input image.

Abstract: Stereo correspondence smoothness tool techniques are described. In one or more implementations, an indication is received of a user-defined region in at least one of a plurality of stereoscopic images of an image scene. Stereo correspondence is calculated of image data of the plurality of stereoscopic images of the image scene, the calculation performed based at least in part on the user-defined region as indicating a smoothness in disparities to be calculated for pixels in the user-defined region.

Abstract: A system and method are disclosed for conserving power during navigation, e.g., user device pointer/cursor navigation, using a fingerprint image sensor, that may comprise processing, via a computing device, fingerprint image sensor data indicative of finger position and movement with respect to a fingerprint image sensor surface in a finger navigation mode to determine if the finger is in a first finger navigation mode; processing, via the computing device, fingerprint image sensor data indicative of finger position and movement with respect to a fingerprint image sensor surface in a finger navigation mode to determine if the finger is in a second finger navigation mode; and transitioning, via the computing device, the fingerprint image sensor from a first power consumption mode to a second power consumption mode, based on detecting a transition from the first finger navigation mode to the second finger navigation mode.

Abstract: A method and apparatus for constructing a map for a mobile robot to be able to reduce a data amount and increase an approach speed. The method includes: searching for a plurality of feature data occupying an arbitrary space by scanning a surrounding environment of the mobile robot; performing quadtree segmentation on first feature data of the plurality of feature data to generate a plurality of first node information as a result of the quadtree segmentation; determining a position of second feature data of the plurality of feature data with respect to the first feature data; and performing a neighborhood moving algorithm for generating a plurality of second node information of the second feature data according to the position of second feature data by using the plurality of first node information.

Abstract: In various example embodiments, a system and method for using machine learning to define user controls for image adjustment is provided. In example embodiments, a new image to be adjusted is received. A weight is applied to reference images of a reference dataset based on a comparison of content of the new image to the reference image of the reference dataset. A plurality of basis styles is generated by applying weighted averages of adjustment parameters corresponding to the weighted reference images to the new image. Each of the plurality of basis styles comprises a version of the new image with an adjustment of at least one image control based on the weighted averages of the adjustment parameters of the reference dataset. The plurality of basis styles is provided to a user interface of a display device.

Abstract: Disclosed are novel techniques for high-precision sex determination and age estimation using facial images. The disclosed age estimation method includes: a step in which facial image data of a subject is acquired; a step in which spatial frequency intensities are calculated from the acquired facial image data; and a step in which the estimated age of the subject is calculated by applying the calculated spatial frequency intensities obtained from the facial image data.

Abstract: Text in web pages or other text documents may be classified based on the images or other objects within the webpage. A system for identifying and classifying text related to an object may identify one or more web pages containing the image or similar images, determine topics from the text of the document, and develop a set of training phrases for a classifier. The classifier may be trained and then used to analyze the text in the documents. The training set may include both positive examples and negative examples of text taken from the set of documents. A positive example may include captions or other elements directly associated with the object, while negative examples may include text taken from the documents, but from a large distance from the object. In some cases, the system may iterate on the classification process to refine the results.

Abstract: Determining a match between the subjects of first and second images as a function of decimal-number representations of regions of the first and second images. The decimal-number representations are generated by performing discrete transforms on the regions so as to obtain discrete-transform coefficients, performing local-bit-pattern encoding of the coefficients to create data streams, and converting the data streams to decimal numbers. In one embodiment, the first and second images depict periocular facial regions, and the disclosed techniques can be used for face recognition, even where a small portion of a person's face is captured in an image. Subspace modeling may be used to improve accuracy.

Abstract: The purpose of the present invention is to provide an image processing apparatus and a computer program such that correspondence points between design data and an edge line or between edge lines can be accurately identified for their matching. In an embodiment for achieving the purpose, when positioning between a first pattern formed by a first line segment and a second pattern formed by a second line segment is performed, a first correspondence point and a second correspondence point are set on the first line segment and the second line segment, respectively; a degree of alignment for performing the positioning of the first pattern and the second pattern is calculated on the basis of the distance between the first correspondence point and the second correspondence point; and the position of the first correspondence point and/or the second correspondence point is changed in accordance with a shape difference between the first line segment and the second line segment (see FIG. 2).

Abstract: A method, system or computer usable program product for authenticating a fingerprint by aggregating samples on a digital touch screen including capturing multiple data samples of a finger swipe across the touch screen, wherein the touch screen is used both for fingerprint recognition and manipulation of a user interface; utilizing a processor to aggregate the multiple data samples to form a combined sample fingerprint of the finger; and responsive to detecting the combined sample fingerprint matching a predetermined allowed fingerprint, providing a signal indicating the fingerprint is authentic.

Abstract: A method and system for detection of video segments in compressed digital video streams is presented. The compressed digital video stream is examine to determine synchronization points, and the compressed video signal is analyzed following detection of the synchronization points to create video fingerprints that are subsequently compared against a library of stored fingerprints.

Abstract: There is provided an image processing device including an image quality configuring part configured to configure image quality on a per-recognizable scene of an image basis according to an instruction from a user, an image quality storage configured to store the image quality configured for each scene, an image recognition part configured to recognize a scene of an image to be acquired, and an image processing part configured to perform image processing on the image based on the image quality configured for the recognized scene by the image quality configuring part.

Abstract: Apparatus for matching a query image against a catalog of images, comprises: a feature extraction unit operative for extracting principle features from said query image; a relationship unit operative for establishing relationships between a given principle feature and other features in the image, and adding said relationships as relationship information alongside said principle features; and a first comparison unit operative for comparing principle features and associated relationship information of said query image with principle features and associated relationship information of images of said catalog to find candidate matches.

Abstract: Systems and methods for analyzing test strip comb members having a plurality of fingers are disclosed. The systems and methods may analyze a test strip comb member to determine the presence of one more analytes on each of the plurality of fingers.

Abstract: A method of observing a three-dimensional image of an examinee's eye to be examined, includes: obtaining a first three-dimensional image which is a three-dimensional image of an anterior segment of the examinee's eye by optical coherence tomography for obtaining a tomographic image by converging a measurement light on the anterior segment of the examinee's eye; obtaining a second three-dimensional image which is a three-dimensional image of a fundus of the examinee's eye by optical coherence tomography for obtaining a tomographic image by converging the measurement light on the fundus of the examinee's eye by a timing different from a timing of obtaining the first three-dimensional image; constructing a three-dimensional eyeball image of the examinee's eye through image processing based on the obtained first and second three-dimensional images; and displaying the constructed three-dimensional eyeball image on a monitor.

Abstract: Various embodiments provide a method for randomly selecting a region on a map for testing and a map of the region can be generated using multiple map rendering engines. A screenshot of each of the generated maps can be obtained and text associated with map labels, such as street, city, and attraction names, can be recognized using an optical character recognition (OCR) engine. At this point, the recognized text from each rendering engine can then be compared to identify at least one error or inconsistency. In at least one embodiment, categories of errors that need most attention in the specific geographic areas can be identified and a human quality assurance tester can isolate these instances and narrow down the same to identify the rendering or data problem.

Abstract: Techniques are disclosed for removing false-positive foreground pixels resulting from environmental illumination effects. The techniques include receiving a foreground image and a background model, and determining an approximated reflectance component of the foreground image based on the foreground image itself and a background model image which is used as a proxy for an illuminance component of the foreground image. Pixels of the foreground image having approximated reflectance values less than a threshold value may be classified as false-positive foreground pixels and removed from the foreground image. Further, the threshold value used may be adjusted based on various factors to account for, e.g., different illumination conditions indoors and outdoors.

Abstract: Techniques are disclosed for analyzing a scene depicted in an input stream of video frames captured by a video camera. In one embodiment, e.g., a machine learning engine may include statistical engines for generating topological feature maps based on observations and a detection module for detecting feature anomalies. The statistical engines may include adaptive resonance theory (ART) networks which cluster observed position-feature characteristics. The statistical engines may further reinforce, decay, merge, and remove clusters. The detection module may calculate a rareness value relative to recurring observations and data in the ART networks. Further, the sensitivity of detection may be adjusted according to the relative importance of recently observed anomalies.

Abstract: To identify a media item from a database of media items that have common content, a region of interest is defined to include a plurality of frames of a test fingerprint that correspond to different segments of a media item. A media identification system queries a database of reference fingerprints to identify candidate reference fingerprints that contain a frame that matches a frame of the test fingerprint. When a candidate reference fingerprint is found, additional matching frames are determined and the region of interest is reduced to eliminate the matched frames of the test fingerprint. This continues until the region of interest is empty or there are no further matching candidates. Once the set of candidate reference fingerprints are identified, the media identification system compares the test fingerprint to the candidates to determine a closest match, thereby identifying the media item associated with the test fingerprint.