Abstract: A panel view module receives a comic comprising one or more pages, which comprise one or more panels. The panel view module may identify candidate comic panels and determine confidence levels for the candidate comic panels. The panel view may also generate a panel view for the comic based on the confidence levels for the candidate comic panels.

Abstract: An information processing apparatus includes a network creating unit that creates a network in which respective characters of plural character recognition results are represented as nodes, and in which nodes of adjacent character images are connected with a link, a first determining unit that determines a first candidate boundary in the network, a second determining unit that determines a second candidate boundary different from the first candidate boundary in the network, and an extracting unit that extracts, as to-be-searched objects, plural candidate character strings from a set of candidate character strings each formed of nodes between the first candidate boundary and the second candidate boundary.

Abstract: Methods for removing vertical and horizontal lines from a document image. The horizontal line removal method includes: for a column of black pixels at each horizontal position along the line, removing them if their maximum stroke width is less than the median value of maximum stroke widths in a small window centered at that horizontal position; Remove connected components remaining in the horizontal line bounding box that do not extend significantly above or below the bounding box boundaries; and Perform closing operation to join broken pieces of character strokes caused by underline removal. This method preserves character strokes while removing underlines. The vertical line removal method includes: for vertical lines that have large height to width ratio, remove parts of such lines that are not at intersection of with horizontal or near-horizontal lines; remove all remaining connected components that touch neither left nor right boundary of the bounding box.

Abstract: Embodiments are directed towards determining within a digital camera whether a pixel belongs to a foreground or background segment within a given image by evaluating a ratio of derivative and deviation metrics in an area around each pixel in the image, or ratios of derivative metrics across a plurality of images. For each pixel within the image, a block of pixels are examined to determine an aggregate relative derivative (ARD) in the block. The ARD is compared to a threshold value to determine whether the pixel is to be assigned in the foreground segment or the background segment. In one embodiment, a single image is used to determine the ARD and the pixel segmentation for that image. Multiple images may also be used to obtain ratios of a numerator of the ARD, useable to determine an extent of the foreground.

Abstract: A method for removing false foreground image content in a foreground detection process performed on a video sequence includes, for each current frame, comparing a feature value of each current pixel against a feature value of a corresponding pixel in a background model. The each current pixel is classified as belonging to one of a candidate foreground image and a background based on the comparing. A first classification image representing the candidate foreground image is generated using the current pixels classified as belonging to the candidate foreground image. The each pixel in the first classification image is classified as belonging to one of a foreground image and a false foreground image using a previously trained classifier. A modified classification image is generated for representing the foreground image using the pixels classified as belonging to the foreground image while the pixels classified as belonging to the false foreground image are removed.

Abstract: Methods and apparatus to count people in images are disclosed. An example method includes generating a first fluctuation factor for a first frame by averaging fluctuation values of a random set of pixels of the first frame; generating a second fluctuation factor for a person indication area of the first frame by averaging fluctuation values of pixels of the person indication area; and marking the person indication area as a false positive when the second fluctuation factor is less than or equal to the first fluctuation factor.

Abstract: The present invention provides a method and an apparatus for detecting traffic video information. The method includes: acquiring a traffic video stream; determining color features of each frame of image in the traffic video stream; calculating the inter-frame distance between adjacent frames according to the color features; calculating the boundary of an image clustered frames' group according to the inter-frame distance by adopting an image clustering evaluation standard in RGB space and an image clustering evaluation standard in YUV space respectively; and determining a final boundary of the image clustered frames' group according to the boundaries of the image clustered frames' group in RGB space and YUV space. By using the present invention, the stability of detection results in different environments may be improved.

Abstract: A system and method is provided where, in one aspect, a user signals that the user is interested in a particular image being displayed as part of a video, wherein objects in the image are detected and recognized, and wherein an action is taken based on information associated with the recognized objects. By way of example only, the actions may include recording a show (with or without a user's prior confirmation) or displaying information about a product in the image.

Abstract: A computer-implemented method of acquiring tax data for use in tax preparation application includes acquiring an image of at least one document containing tax data therein with an imaging device. A computer extracts one or more features from the acquired image of the at least one document and compares the extracted one or more features to a database containing a plurality of different tax forms. The database may include a textual database and/or geometric database. The computer identifies a tax form corresponding to the at least one document from the plurality of different tax forms based at least in part on a confidence level associated with the comparison of the extracted one or more features to the database. At least a portion of the tax data from the acquired image is transferred into corresponding fields of the tax preparation application.

Abstract: Systems and methods are provided for managing depths in stereo images. In one implementation, systems and methods are provided for using a procedural shader and automated rendering to match depths of computer-generated images with that of native stereo images, i.e., images recorded using two cameras for stereo imaging. The system and method may employ a checkerboard and a tri-planar shader, and may further employ a texture reference object to make more convenient and intuitive the depth matching.

Abstract: According to one embodiment, approximate named-entity extraction from a dictionary that includes entries is provided, where each of the entries includes one or more words. Words are read from the entries of the dictionary, and network resources are searched to determine a frequency of occurrence of the words on the network resources. In view of the frequency of occurrence of the words located on the network resources, domain relevancy of the words in the entries of the dictionary is determined. A domain repository is created using top-ranked words as determined by the domain relevancy of the words. In view of the domain repository, signatures for both the entries of the dictionary and strings of an input document are computed. The strings of the input document are filtered by comparing the signatures of the strings against the signatures of the entries to identify approximate-match entity names.

Abstract: A computer-implemented method for determining a subject's fitment for a garment, has the steps of receiving a body type, receiving a body stance, receiving two or more photos of the subject comprising at least a side view photo and a front or rear view photo, determining the scale of each of the two or more photos, determining certain measurements from the two or more photos using the scale, and calculating further measurements from the determined measurements. The method may have the further steps of receiving one or more measurements comprising at least a height from the subject, and using the one or more measurements from the subject to determine a scale of each of the two or more photos. Alternatively, the scale of each of the two or more photos may be determined by a depth-sensing camera.

Abstract: Techniques for semantically annotating images in a plurality of images, each image in the plurality of images comprising at least one image region. The techniques include identifying at least two similar images including a first image and a second image, identifying corresponding image regions in the first image and the second image, and assigning, using at least one processor, annotations to image regions in one or more images in the plurality of images by using a metric of fit indicative of a degree of match between the assigned annotations and the corresponding image regions. The metric of fit may depend on at least one annotation for each image in a subset of the plurality of images and the identified correspondence between image regions in the first image and the second image.

Abstract: Embodiments of apparatuses, systems, and methods for a temporal hole filling are described. Specifically, an embodiment of the present invention may include a depth-based hole filling process that includes a background modeling technique (SGM). Beneficially, in such an embodiment, holes in the synthesized view may be filled effectively and efficiently.

Abstract: The present invention relates to an image processor that is configured to compare a single frame included in the received signal with a preceding processed frame so as to detect changes. In order to enable the image being compressed in as compact a format as possible, the processor is configured to detect changes by dividing the pixels of a single frame into movement blocks and to detect the movement block as changed if in said movement block there is a given number of pixels whose color value change as compared with the corresponding pixels of the preceding processed frame exceeds a predetermined threshold, to exchange the color values of the pixels of the unchanged movement blocks in the frame under process for predetermined values and to generate a signal that indicates the changes.

Abstract: A system for arranging searchable items, such as digital images or multimedia icons for presentation on a graphical user interface. The system has an optimizer module which minimizes a cost function for arranging items in a layout space and is applied to one or more predetermined characteristics of the items. The optimizer module also creates a mixture distribution by considering each item as having a spatial distribution in the layout space and controls the entropy of the mixture distribution in order to maximize the extent to which each item occupies a separate position in the layout space. A Renderer module then renders the layouts to generate a display.

Abstract: In an image processing apparatus, an image pickup unit takes images of an object including the face of a person wearing the glasses by which to observe a stereoscopic image that contains a first parallax image and a second parallax image obtained when the object in a three-dimensional (3D) space is viewed from different viewpoints. A glasses identifying unit identifies the glasses included in the image of the object taken by the image pickup unit. A face detector detects a facial region the face of the person included in the image of the object taken by the image pickup unit, based on the glasses identified by the glasses identifying unit. An augmented-reality special rendering unit adds a virtual feature to the facial region of the face of the person detected by the face detector.

Abstract: Embodiments of the present invention are directed to a method and system for altering a presentation of broadcast content. More specifically, embodiments provide an accurate and efficient mechanism for suppressing advertisements by using downloaded templates to identify advertisements before suppression for a duration indicated by information downloaded about the advertisement, where the suppression is effectuated by controlling a display or set-top box using remote-control functionality. In one embodiment, a digital signal comprising broadcast content is accessed. A plurality of templates are received and stored, wherein each template represents a known portion of the broadcast content. The digital signal is then compared against a portion of the plurality of stored templates. Upon a match, a control signal operable to alter a component of a presentation of the known portion of the broadcast content is generated.

Abstract: The present invention relates to a video image processor that is configured to compare a single frame included in the received video signal with a preceding processed frame so as to detect changes. In order to enable the video being compressed in as compact a format as possible the processor is configured to detect changes by dividing the pixels of a single frame into movement blocks and to detect the movement block as changed if in said movement block there is a given number of pixels whose colour value change as compared with the corresponding pixels of the preceding processed frame exceeds a predetermined threshold, to exchange the colour values of the pixels of the unchanged movement blocks in the frame under process for predetermined values and to generate a signal that indicates the changes.

Abstract: One exemplary embodiment involves receiving, at a computing device comprising a processor, a test image having a candidate object and a set of object images detected to depict a similar object as the test image. The embodiment involves localizing the object depicted in each one of the object images based on the candidate object in the test image to determine a location of the object in each respective object image and then generating a validation score for the candidate object in the test image based at least in part on the determined location of the object in the respective object image and known location of the object in the same respective object image. The embodiment also involves computing a final detection score for the candidate object based on the validation score that indicates a confidence level that the object in the test image is located as indicated by the candidate object.

Abstract: The invention involves a method for processing of machine-readable forms or documents of non-fixed format. The method makes use of, for example, a structural description of characteristics of document elements, a description of a logical structure of the document, and methods of searching for document elements by using the structural description. A structural description of the spatial and parametric characteristics of document elements and the logical connections between elements may include a hierarchical logical structure of the elements, specification of an algorithm of determining the search constraints, specification of characteristics of every searched element, and specification of a set of parameters for a compound element identified on the basis of the aggregate of its components. The method of describing the logical structure of a document and methods of searching for elements of a document may be based on the use of the structural description.

Abstract: An improved method for generating high dynamic range images by modifying the weight function used in conventional methods to weigh pixel values when combining multiple images in an input image set. Different weight functions, as functions of intensity, are used for different images (brackets). For darker brackets, a region near the high end of the intensity range is given higher weight values than a symmetrical region near the lower end of the intensity range; for brighter brackets, a region near the lower end of the intensity range is given higher weight values than a symmetrical region near the higher end of the intensity range. The weight function for the middle brackets are kept symmetrical but its function is non-zero at the two end points of the intensity range. This is effective for reducing chromatic artifacts for under- or over-exposed areas, in particular when ghost artifact removal is incorporated.

Abstract: Systems and methods for image segmentation using a deformable atlas are provided. One method includes obtaining one or more target images, obtaining one or more propagated label probabilities for the one or more target images, and segmenting the one or more target images using a cost function of a deformable atlas model. The method further includes identifying segmented structures within the one or more target images based on the segmented one or more target images.

Abstract: Systems and methods for implementing a hierarchical image recognition framework for classifying digital images are provided. The provided hierarchical image recognition framework utilizes a multi-layer approach to model training and image classification tasks. A first layer of the hierarchical image recognition framework generates first layer confidence scores, which are utilized by the second layer to produce a final recognition score. The provided hierarchical image recognition framework permits model training and image classification tasks to be performed more accurately and in a less resource intensive fashion than conventional single-layer image recognition frameworks. In some embodiments real-time operator guidance is provided for an image classification task.

Abstract: Provided is a noise-removing system including: an iris controller configured to control a degree of openness of a lens iris; a noise image analyzer configured to perform an analysis of a noise image that is obtained in a state when the iris controller closes the iris; a normal image analyzer configured to analyze a normal image that is obtained in a state when the iris controller opens the iris; and a noise remover configured to remove noise from an image, based on a result of the analysis performed by the noise image analyzer and the analysis performed by the normal image analyzer.

Abstract: The object detection apparatus prevents or eliminates detection errors caused by changes of an object which frequently appears in a background. To this end, an object detection apparatus includes a detection unit which detects an object region by comparing an input video from a video input device and a background model, a selection unit which selects a region of a background object originally included in a video, a generation unit which generates background object feature information based on features included in the background object region, and a determination unit which determines whether or not the object region detected from the input video is a background object using the background object feature information.

Abstract: A method for processing data includes receiving a depth map of a scene containing a humanoid form. The depth map is processed so as to identify three-dimensional (3D) connected components in the scene, each connected component including a set of the pixels that are mutually adjacent and have mutually-adjacent depth values. Separate, first and second connected components are identified as both belonging to the humanoid form, and a representation of the humanoid form is generated including both of the first and second connected components.

Abstract: A method and system for detecting a motion of a target object in a thermal image by removing a shadow by heat of the target object from the thermal image. The motion detecting system includes: a learning unit obtaining at least one of size and brightness of a shadow by heat of a reference object based on characteristics of the shadow by heat of the reference object by temperature; and a detecting unit removing a shadow region of the target object from the thermal image including the target object based on at least one of the size and the brightness of the shadow by heat of the object.

Abstract: An information processing apparatus encodes an input pattern to a code including a plurality of bits, calculates reliabilities for respective bits of the code, generates a similar codes each similar to the code based on the reliabilities, and recognizes the input pattern based on the code and the similar codes.

Abstract: An apparatus for 3D representation of image data, comprising: a structure identifier for identifying structures in motion within image data, and a skeleton insertion unit, which associates three-dimensional skeleton elements with the identified structures. The skeleton elements are able to move with the structures to provide a three-dimensional motion and structural understanding of said image data which can be projected back onto the input data. As well as individual elements, complex bodies can be modeled by complex skeletons having multiple elements. The skeleton elements themselves can be used to identify the complex objects.

Abstract: What is disclosed is a system and method for selecting the optimal wavelength ban combination for a multi-band infrared camera system which is optimized for skin detection. An objective function is constructed specifically for this application from classified performance and the algorithm generates wavelengths by maximizing the objective function. A specific wavelength band combination is selected which maximizes the objective function. Also disclosed is a 3-band and 4-band camera system with filters each having a transmittance of one of a combination of wavelength bands optimized to detect skin in the infrared band. The camera systems disclosed herein find their intended uses in a wide array of vehicle occupancy detection systems and applications. Various embodiments are disclosed.

Abstract: A system of image stereo matching includes at least one stereo matching unit (SMU) each receives a first view and a second view of a view pair, according to which the SMU generates a first depth map for the first view. The system also includes a backward tracer operable to receive the first depth map, according to which a second depth map for the second view is derived.

Abstract: A system for deploying solar panels may include a classification module configured to receive an image of a structure with a roof and to classify portions of the image that include portions of the roof as roof image segments. The system may also include an identification module configured to identify a property of the roof based at least in part on the roof image segments and a configuration module configured to determine a configuration of one or more solar panels for placement on the roof based on the identified property of the roof.

Abstract: A method and a graphical user interface for processing of an object in light field data is provided. The method and graphical user interface may be used to remove an object or obstruction blocking another object or background of a captured scene. The method and graphical user interface may also be used for high precision edge detection in a light field image.

Abstract: A method of segmenting an image into halftone and non-halftone regions includes: processing a pixel block of a grayscale image using a thresholding algorithm and an intensity threshold to form a binary block, the pixel block including grayscale pixels with intensity values, the binary block including binary pixels with binary values; identifying a pixel group in the binary block, wherein the pixel group includes connected pixels represented by a select binary value, wherein pixels in the pixel group adjacent to pixels of the pixel block represented by another binary value are identified as boundary pixels of the pixel group; and processing the pixel group using a scale invariant feature transform (SIFT) algorithm to extract keypoints from the pixel group to form a keypoint map in conjunction with segmenting the grayscale image into halftone and non-halftone regions. An image processing device for performing the method is also provided.

Abstract: Techniques for generating cross-modality semantic classifiers and using those cross-modality semantic classifiers for ground level photo geo-location using digital elevation are provided. In one aspect, a method for generating cross-modality semantic classifiers is provided. The method includes the steps of: (a) using Geographic Information Service (GIS) data to label satellite images; (b) using the satellite images labeled with the GIS data as training data to generate semantic classifiers for a satellite modality; (c) using the GIS data to label Global Positioning System (GPS) tagged ground level photos; (d) using the GPS tagged ground level photos labeled with the GIS data as training data to generate semantic classifiers for a ground level photo modality, wherein the semantic classifiers for the satellite modality and the ground level photo modality are the cross-modality semantic classifiers.

Abstract: A radius parameter may be included with mapping data requests resulting from “longpress” fingertip selection actions on client computing devices including touch screen display interfaces. The radius parameter may be based on the approximate size of a user's finger in pixels. Based on a current zoom level and/or altitude and a latitude, the radius may be used with a finger tip size to determine how far away a location on the screen from the finger tip center is. A remote mapping server may use the radius as a distance limit for performing a snapping action. For example, if the best candidate for a feature snap is outside the range of the longpress radius, then the feature snap may be abandoned, and the remote mapping server returns a dropped pin response to the client. The user may then customize the dropped pin for sharing or other actions.

Abstract: A method for processing an image of a scene of interest includes receiving an original target image of a scene of interest at an image processing device from an image source device, the original target image exhibiting shadowing effects associated with the scene of interest when the original target image was captured, the original target image comprising a plurality of elements and representing an instantaneous state for the scene of interest, pre-processing the original target image using a modification identification algorithm to identify elements of the original target image to be modified, and generating a copy mask with a mask region representing the elements to be modified and a non-mask region representing other elements of the original target image. An image processing device for processing an image of a scene of interest and a non-transitory computer-readable medium are also provided.

Abstract: Various arrangements for modeling a scene are presented. A plurality of images of the scene captured over a period of time may be received, each image comprising a plurality of pixels. A plurality of background models may be created using the plurality of images. At least one background model may be created for each pixel of the plurality of pixels. A plurality of foreground models may be created using the plurality of images. A foreground model may be created for each pixel of at least a first subset of pixels of the plurality of pixels. The background models and the foreground models may be indicative of the scene over the period of time.

Abstract: Extracting financial card information with relaxed alignment comprises a method to receive an image of a card, determine one or more edge finder zones in locations of the image, and identify lines in the one or more edge finder zones. The method further identifies one or more quadrilaterals formed by intersections of extrapolations of the identified lines, determines an aspect ratio of the one or more quadrilateral, and compares the determined aspect ratios of the quadrilateral to an expected aspect ratio. The method then identifies a quadrilateral that matches the expected aspect ratio and performs an optical character recognition algorithm on the rectified model. A similar method is performed on multiple cards in an image. The results of the analysis of each of the cards are compared to improve accuracy of the data.

Abstract: A method is provided for dividing a pattern into a plurality of sub-patterns, each sub-pattern being adapted for use with an image search method that can provide a plurality of sub-pattern search results. The method represents the pattern as a plurality of feature points, generates candidate partitions of the plurality of feature points, and then scores the candidate partitions by examining characteristics of each potential sub-pattern of each candidate partition. The highest-scoring partition is selected, and then it is applied to the plurality of feature points, creating one or more sub-pluralities of features. The invention advantageously provides a plurality of sub-patterns where each sub-pattern contains enough information to be located with a feature-based search method, where that information has been pre-evaluated as being useful and particularly adapted for running feature-based searches.

Abstract: An image processing apparatus groups distance measurement areas for every group of distances close to each other based on a photographing distance obtained for each distance measurement area and assigns the same distance number to the distance measurement areas belonging to the same group. The image processing apparatus further divides a photographed image into a plurality of blocks to calculate an average value of hue of each block and sets a plurality of blocks having similar average values of hue among the blocks that are adjacent to one another as the same (one) color area. The image processing apparatus then divides the color area based on the distance number overlapping with the color area.

Abstract: A method and system for real time processing of a sequence of video frames. A current frame in the sequence and at least one frame in the sequence occurring prior to the current frame is analyzed. The sequence of video frames is received in synchronization with a recording of the video frames in real time. The analyzing includes performing a background subtraction on the at least one frame, which determines a background image and a static region mask associated with a static region consisting of a contiguous distribution of pixels in the current frame, which includes executing a mixture of 3 to 5 Gaussians algorithm coupled together in a linear combination by Gaussian weight coefficients to generate the background model, a foreground image, and the static region. The static region mask identifies each pixel in the static region upon the static region mask being superimposed on the current frame.

Abstract: Sharing a desktop screen is implemented on a Hypertext Transfer Protocol (HTTP) web server using variable interval GET and POST requests. Connection to a web server bypasses firewall or shared IP limitations. In the described HTTP/HTTPS only solution, an existing web server is used to allow one or more clients to view a remote computer's screen and control it (remoting). The web server can act as an intermediary to enable one computer to “take control” of a second computer by enabling the first computer (a client) to view the desktop session of the second computer (a host) and enable the first computer to act as master for input device input (e.g., keyboard input, mouse input, stylus input, etc.) for the second computer.

Abstract: The disclosure describes novel technology for inferring scenes from images. In one example, the technology includes a system that can determine partition regions from one or more factors that are independent of the image data, for an image depicting a scene; receive image data including pixels forming the image; classify pixels of the image into one or more pixel types based on one or more pixel-level features; determine, for each partition region, a set of pixel characteristic data describing a portion of the image included in the partition region based on the one or more pixel types of pixels in the partition region; and classify a scene of the image based on the set of pixel characteristic data of each of the partition regions.

Abstract: A method and system for real time processing of a sequence of video frames. A current frame in the sequence and at least one frame in the sequence occurring prior to the current frame is analyzed. The sequence of video frames is received in synchronization with a recording of the video frames in real time. The analyzing includes performing a background subtraction on the at least one frame, which determines a background image and a static region mask associated with a static region consisting of a contiguous distribution of pixels in the current frame. The static region mask identifies each pixel in the static region upon the static region mask being superimposed on the current frame. A status of a static object is determined as either an abandoned status if the static object is an abandoned object or a removed status if the static object is a removed object.

Abstract: A method and an apparatus for bilayer image segmentation are described. A set of segmentation seeds for the image is generated by analyzing a depth histogram of the image. Then a segmentation map is generated by minimizing an objective function, which models a directed flow from the foreground segmentation seeds towards the background segmentation seeds.

Abstract: Disclosed herein are techniques for performing automated structure delineation on image data using trained landmark detectors and a shape refinement tool. The landmark detectors can be trained to detect a landmark in the image data based on image features that are indicative of intensity variations over a plurality of windows of the image data points. A machine-learning algorithm can be used to train the landmark detectors. The landmarks in the image data that are detected by the trained landmark detects can be used to initialize an iterative shape refinement to thereby compute a refined shape estimate for a structure of interest such as a prostate.

Abstract: Multi-mode video event indexing includes determining a quality of object distinctiveness with respect to images from a video stream input. A high-quality analytic mode is selected from multiple modes and applied to video input images via a hardware device to determine object activity within the video input images if the determined level of detected quality of object distinctiveness meets a threshold level of quality, else a low-quality analytic mode is selected and applied to the video input images via a hardware device to determine object activity within the video input images, wherein the low-quality analytic mode is different from the high-quality analytic mode.

Abstract: A method and system for real time processing of a sequence of video frames. A current frame in the sequence and at least one frame in the sequence occurring prior to the current frame is analyzed. Each frame includes a two-dimensional array of pixels. The sequence of video frames is received in synchronization with a recording of the video frames in real time. The analyzing includes performing a background subtraction on the at least one frame, which determines a background image and a static region mask associated with a static region consisting of a contiguous distribution of pixels in the current frame. The static region mask identifies each pixel in the static region upon the static region mask being superimposed on the current frame. A determination is made that a persistence requirement, a non-persistence duration requirement, and a persistence duration requirement have been satisfied.