Abstract: A three-dimensional image processing apparatus includes an obtainer that obtains three-dimensional image information including information of a first image and a second image, a shade information obtainer that obtains shade information from the information of the first image and/or the second image, and a disparity adjuster that adjusts a disparity of a subject contained in the first and the second images based on the shade information.

Abstract: Described is a method and system for embedding unsupervised learning into three critical processing stages of the spatio-temporal visual stream. The system first receives input video comprising input video pixels representing at least one action and at least one object having a location. Microactions are generated from the input image using a set of motion sensitive filters. A relationship between the input video pixels and the microactions is then learned, and a set of spatio-temporal concepts is learned from the microactions. The system then learns to acquire new knowledge from the spatio-temporal concepts using mental imagery processes. Finally, a visual output is presented to a user based on the learned set of spatio-temporal concepts and the new knowledge to aid the user in visually comprehending the at least one action in the input video.

Abstract: A moving object display region where a moving object is displayed and a background display region are extracted based on a difference image of frame images. Blurring processing such as averaging processing is performed on an image in the background display region. Processing for further sharpening an image, such as contrast enhancement, is performed on an image in the moving object display region.

Abstract: An interactive user-friendly incremental calibration technique that provides immediate feedback to the user when aligning a point on a 3D model to a point on a 2D image. A can drag-and-drop points on a 3D model to points on a 2D image. As the user drags the correspondences, the application updates current estimates of where the camera would need to be to match the correspondences. The 2D and 3D images can be overlayed on each other and are sufficiently transparent for visual alignment. The user can fade between the 2D/3D views providing immediate feedback as to the improvements in alignment. The user can begin with a rough estimate of camera orientation and then progress to more granular parameters such as estimates for focal length, etc., to arrive at the desired alignment. While one parameter is adjustable, other parameters are fixed allowing for user adjustment of one parameter at a time.

Abstract: Statistical analysis techniques based on auto- and cross-correlations/cumulants, of image stacks of fluctuating objects are used to improve resolution beyond the classical diffraction limit and to reduce the background. The time trajectory of every pixel in the image frame is correlated with itself and/or with the time trajectory of an adjacent pixel. The amplitude of these auto- or cross-correlations/cumulants of each pixel, at a given time lag or averaged or integrated over an interval of time lags, is used as the intensity value of that pixel in the generated superresolved optical fluctuation image.

Abstract: A high speed filtering apparatus and a method for high precision restoration of a depth image are provided. The high speed filtering apparatus for high precision restoration of the depth image may include a block setting unit to set a first block including a target pixel, and to set a second block with respect to a central pixel distributed around the target pixel based on a size of the first block, a weight determining unit to determine a pixel weight with respect to each pixel in the second block, and to determine a block weight with respect to the second block by applying the pixel weight, and a processor to filter the target pixel based on the block weight, thereby accurately filtering the target pixel.

Abstract: A three-dimensional distance measurement apparatus comprising: a distance information calculation unit configured to calculate distance information of a measurement object based on a captured image of the measurement object on which pattern light is projected; and a local pattern setting unit configured to adaptively set a spatial resolution of measurement points of the pattern light for each local region, wherein the local pattern setting unit sets the spatial resolution of the measurement points of the pattern light based on information about an arithmetic amount of the distance information calculation unit.

Abstract: A color interpolation apparatus includes a threshold value determining unit that calculates a luminance value of each area set based on each pixel of a Bayer pattern image and determines edge determining threshold values, an edge information calculating unit that calculates pixel value variations in multiple directions of the set areas by using pixel values of pixels included in each area and edge information regarding each area by using the multiple directional pixel value variations, an area determining unit that determines an edge type by using the edge determining threshold values and the edge information, and a color interpolation unit that applies a color interpolation filter previously set according to the edge type to a corresponding area according to the edge type. The area determining unit uses green information or interpolated green information of a central pixel depending on whether or not a color of the central pixel is green.

Abstract: A method of storing a content of a three-dimensional image includes a processor initializing a register and a maximum number; the processor utilizing a stereo comparison algorithm to generate a depth information map corresponding to each frame of a three-dimensional image signal; the processor obtaining a depth value corresponding to each pixel of each pixel row of the depth information map from the depth information map; the processor generating at least one depth vector corresponding to the pixel row according to a depth value corresponding to each pixel of the pixel row; a counter counting a number of the at least one depth vector; the processor storing the number of the at least one depth vector in the register; the processor comparing the number of the at least one depth vector with the maximum number; and the processor executing a corresponding operation on the register according to a comparison result.

Abstract: Image data may be processed via one or more stages by a hardware image sensor pipeline (ISP) wherein one or more software processing steps may be inserted at any point within the hardware ISP. Output from any stage of the hardware ISP may be stored in memory. Stored hardware ISP output may be retrieved from memory and processed via one or more software processes. Results from the one or more software processes may be stored in memory and communicated to any stage of the hardware ISP for additional processing. In this regard, the hardware ISP and one or more processors may simultaneously process portions of image data. In addition, the hardware ISP and the one or more processors may be integrated within a chip.

Abstract: OCR errors are identified and corrected through learning. An error probability estimator is trained using ground truths to learn error probability estimation. Multiple OCR engines process a text image, and convert it into texts. The error probability estimator compares the outcomes of the multiple OCR engines for mismatches, and determines an error probability for each of the mismatches. If the error probability of a mismatch exceeds an error probability threshold, a suspect is generated and grouped together with similar suspects in a cluster. A question for the cluster is generated and rendered to a human operator for answering. The answer from the human operator is then applied to all suspects in the cluster to correct OCR errors in the resulting text. The answer is also used to further train the error probability estimator.

Abstract: A method, an apparatus and an article of manufacture for generating a cardiovascular measurement for individual identification. The method includes acquiring at least one depiction of cardiac anatomy from an individual, extracting at least one quantified representation of cardiac anatomy from the at least one depiction, defining at least one comparison technique between the at least one quantified representation of cardiac anatomy and at least one additional quantified representation of cardiac anatomy, and identifying the individual based on the at least one defined comparison technique.

Abstract: A server is configured to identify items of visual content; assign the identified items to clusters; assign the identified items, of a particular cluster, to layers; generate a document that includes information regarding the identified items arranged according to the clusters and the layers; and provide the document. The document may include, for the particular cluster, a first item, corresponding to one of the identified items assigned to a first layer, and second items, corresponding to multiple ones of the identified items assigned to a second layer. The size of the first item may be greater than a size of each of the second items. The first item may be presented on top of the second items, within the document, to at least partially cover the second items, where at least one of the second items is at least partially visible within the document.

Abstract: Provided are a striped pattern image examination support device, method, and program. The device includes: image transformation element for transforming at least one of two striped pattern images so as to cause coordinates of charting points, which are points that correspond across the two striped pattern images, to match in a plurality of pairs of the charting points which are included in the two striped pattern images; intersecting point extraction element for calculating coordinates of intersecting points of stripes in the striped pattern images and line segments each of which connects two of the charting points in the striped pattern images; charting diagram display element for displaying the two striped pattern images after transformation by the image transformation element, and displaying figures representing charting points at positions corresponding to coordinates after the transformations of each of the charting points on the two striped pattern images.

Abstract: An image processing method capable of appropriately increasing a resolution of an input image in which an edge direction that is a direction along an edge included in the input image is identified; a shape of an application region that is a region including at least a part of the edge is determined according to the identified edge direction; an image similar to an image within the application region having the determined shape is searched for; and an output image is generated by performing a resolution conversion process on the input image using the similar image so that the input image includes high-frequency components.

Abstract: A defect image processing apparatus uses a normalized cross correlation to image-match a layout image (52) acquired from a design data with an image acquired by removing, from a defect image (53), the defect area portions thereof, and displays, as a result of that matching, a layout image and defect image (54) on the display device. In the displayed layout image & defect image (54), not only the layout image, the layer of which is the same as that of the defect image (53), but also a layout image of another layer is displayed superimposed on the defect image (53). This makes it easier to analyze the factor of a systematic defect having occurred due to a positional relationship with another layer.

Abstract: Systems and methods for processing image signals are described. One method comprises obtaining a generator signal based on an image signal and determining relative latencies associated with two or more pulses in a pulsed signal using a function of the generator signal that can comprise a logarithmic function. The function of the generator signal can be the absolute value of its argument. Information can be encoded in the pattern of relative latencies. Latencies can be determined using a scaling parameter that is calculated from a history of the image signal. The pulsed signal is typically received from a plurality of channels and the scaling parameter corresponds to at least one of the channels. The scaling parameter may be adaptively calculated such that the latency of the next pulse falls within one or more of a desired interval and an optimal interval.

Abstract: The claimed subject matter describes a novel technique to measure the beam profile using an area detector. In one embodiment, a set of one-dimensional beam profile measurements is performed by taking two images under the same source conditions but at two different positions of the detector, with each position of the detector shifted by a certain distance in the direction corresponding to the direction of the one-dimensional profile measurement. In further embodiments, a set of two-dimensional beam profile measurements is achieved by determining a second set of one-dimensional profiles from the same sampling points in a second direction and building a two-dimensional map of the beam profile by correlating the first one-dimensional profile measurement with the second one-dimensional profile measurement.

Abstract: A method and apparatus to generate an object descriptor using extended curvature gabor filters. The method and apparatus may increase a recognition rate of even a relatively small image with use of an extended number of curvature gabor filters having controllable curvatures and may reduce the amount of calculation required for face recognition by performing the face recognition using only some of the extended curvature gabor filters which have a great effect on the recognition rate. The object descriptor generating method includes extracting gabor features from an input object image by applying a plurality of curvature gabor filters, generated via combination of a plurality of curvatures and a plurality of Gaussian magnitudes, to the object image, and generating an object descriptor for object recognition by projecting the extracted features onto a predetermined base vector.

Abstract: An input image (7) having a first pixel resolution is acquired from an image capture system (2). A respective characterization of each of at least one visual quality feature of the input image (7) is determined. An output thumbnail image (9) is produced from the input image (7). The output thumbnail image (9) reflects the respective characterization of each visual quality feature. The output thumbnail image (9) has a second pixel resolution lower than the first pixel resolution. The output thumbnail image (9) is output in association with operation of the image capture system (2).