Abstract: Devices and techniques are generally described for segmentation of object image data from background image data. In some examples, the segmentation of object image data may comprise capturing image data comprising color data and depth data. In some examples, the segmentation of object image data may further include separating the depth data into a plurality of clusters of image data, wherein each cluster is associated with a respective range of depth values. In various examples, the segmentation of object image data may comprise selecting a main cluster of image data as corresponding to an object of interest in the image data. In various other examples, the segmentation of object image data may comprise identifying pixels of the main cluster that correspond to the object of interest.

Abstract: Systems, methods, and computer-readable media to improve multi-image color-refinement operations are disclosed for refining color differences between images in a multi-image camera system with application to disparity estimation. Recognizing that corresponding pixels between two (or more) images of a scene should have not only the same spatial location, but the same color, can be used to improve the spatial alignment of two (or more) such images and the generation of improved disparity maps. After making an initial disparity estimation and using it to align the images, colors in one image may be refined toward that of another image. (The image being color corrected may be either the reference image or the image(s) being registered with the reference image.) Repeating this process in an iterative manner allows improved spatial alignment between the images and the generation of superior disparity maps between the two (or more) images.

Abstract: Server and client methods and systems for improving efficiency, accuracy and speed for inputting data from a variety of networked resources into an electronic form in a continuously streaming manner by multiple operators. More specifically, the present disclosure relates to client/server system and methods for continuous streaming to a series of networked input devices a re-organized forms to allow for multiple operator input to improve speed, accuracy and efficiency of electronic form population.

Abstract: Video data stream storage comprising: receiving, by at least one processor, at least one raw video stream that has at least one object, recording, in at least one memory the at least one raw video stream, extracting, by the at least one processor, at least one object characteristic of the at least one object and at least one timestamp associated with the at least one object characteristic and recording, in the at least one memory the at least one object characteristic and the at least one timestamp associated with the at least one object characteristic.

Abstract: A spherical content capture system captures spherical video content. A spherical video sharing platform enables users to share the captured spherical content and enables users to access spherical content shared by other users. In one embodiment, captured metadata or video/audio processing is used to identify content relevant to a particular user based on time and location information. The platform can then generate an output video from one or more shared spherical content files relevant to the user. The output video may include a non-spherical reduced field of view such as those commonly associated with conventional camera systems. Particularly, relevant sub-frames having a reduced field of view may be extracted from each frame of spherical video to generate an output video that tracks a particular individual or object of interest.

Abstract: Systems and methods for determining an objective metric for analyzing health of a patient's liver are described. In some embodiments, the system may include a scanner that can detect radiation counts responsive to administration of radioactive compound to a patient. Further, the system may include an image detection module that can access image data responsive to the detected radiation counts by the scanner. The image detection module can programmatically identify a first region of interest corresponding to a liver of the patient from the image data. A parameter calculator module can programmatically determine a first attribute associated with the first region of interest and calculate a first parameter indicating health of the liver of the patient based at least in part on the first attribute associated with the first region of interest.

Abstract: A method, apparatus and computer program product provide improved image processing techniques. An example of a method includes receiving a source image, programmatically identifying a plurality of anatomical elements within the source image through use of a computer vision technique, determining a first region of the source image corresponding to a first anatomical element, determining a second region of the source image corresponding to a second anatomical element, applying at least one first configuration setting to the first region, applying at least one second configuration setting to the second region, the at least one second configuration setting different from the at least one first configuration setting, and generating a merged image, wherein the merged image comprises the first region as visualized according to the at least one first configuration setting and the second region as visualized according to the at least one second configuration setting.

Abstract: A visualization system for displaying volumetric renderings includes a display, a user input actuator that is movable over a first range, a second range, and a third range, and a processor. The second range is located between the first range and the third range. The processor is configured to generate a first volumetric rendering according to a first transfer function when the user input actuator is positioned in the first range, and generate the first volumetric rendering according to a second transfer function when the user input actuator is positioned in the third range. The second transfer function is different than the first transfer function. The processor is further configured to generate the first volumetric rendering according to a blending of the first transfer function and the second transfer function when the user input actuator is positioned in the second range, and display the first volumetric rendering on the display.

Abstract: Multi-video registration for video synthesis is described. In example implementations, at least one computing device synthesizes multiple videos to create merged images using an automated mechanism to register the multiple videos. The computing device obtains multiple videos with each video including a sequence of multiple frames. Using multiple camera poses determined in a three-dimensional scene reconstruction, respective frames of respective ones of the multiple videos are linked to produce linked frames. The computing device aligns the linked frames to produce aligned frames using point guidance that is based on the multiple spatial points identified in the 3D scene reconstruction. For example, pixels in each of the linked frames that correspond to a same spatial point of the three-dimensional scene reconstruction can be used to align the linked frames at a pixel level. Based on the aligned frames, the computing device creates at least one merged image to synthesize the multiple videos.

Abstract: An image processing apparatus includes: a controller configured to perform: acquiring objective image data representing an objective image, the objective image data being obtained by optically reading a manuscript; analyzing the objective image data; generating an image processing command in which image processing to be executed for the objective image data is described, by using a result of analysis of the objective image data; and generating an image file including the objective image data and the image processing command. The image processing command is to be executed when a reproduction image is reproduced at a reproduction unit by using the image file.

Abstract: Methods, systems, and computer program products for static image segmentation are provided herein. A method includes segmenting an image containing a target object into multiple regions; analyzing video content containing the target object to determine a similarity metric across the multiple segmented regions based on information associated with the multiple segmented regions; and applying the similarity metric to the image to identify two or more of the multiple segmented regions as being portions of the target object.

Abstract: A system and method for generating a mixed raster content representation of an input image. An input image is segmented into image and text layers. Connected component analysis is performed on the image layer, with each group of connected pixels labeled. For each group, an average color is determined and when a mask layer exists for the color, the pixels in the group are enabled in the mask layer. When a mask layer does not exist for the color, a new mask layer is created and the corresponding pixels enabled. The image layer is then removed and the mask layers are combined into the text layer, whereupon a text only MRC compression file is output.

Abstract: A social media identity/construct is created for a person by locating social media profiles of the person at least by one or both of i) recognizing image(s) in one or more social media postings related to the person and ii) correlating geolocation data embedded in one or more social media postings related to the person or the person's social media connections, analyzing, monitor social media outlets and channels to create a social media profile for the person.

Abstract: An image processing apparatus includes an acquiring unit, and determining unit. The acquiring unit acquires captured images from an imaging device mounted on a vehicle. The determining unit determines the normality of the imaging direction of the imaging device in accordance with whether an appearance pattern of a combination of a portion in which a variation is detected and a portion in which a variation is not detected among the acquired captured images corresponds to a predetermined pattern.

Abstract: In general, this disclosure describes media stream transmission techniques for a field computing device and a central computing device. The central computing device may capture an image of a local background environment. The central computing device may record a first media stream that includes at least a portion of the image of the background environment and at least one movement of at least one object through the background environment. The central computing device may remove the image of the background environment from the first media stream to create a second media stream that includes the movement of the object through the background environment without the image of the background environment. The second media stream may also include timing information that can be synchronized to timing information for a media stream originating at the field computing device such that the second media stream may be superimposed on the field media stream.

Abstract: An image processing apparatus that acquires a focusing distance from a plurality of images each having different degrees of blur, comprises a misalignment detection unit that detects misalignment among the plurality of images; a distance information acquisition unit that acquires, based on the plurality of images, distance information, which is information to indicate a focusing distance in an area of the images; and a reliability information acquisition unit that creates, based on the detected misalignment, reliability information, which is information to indicate reliability of the focusing distance acquired for each area, wherein the distance information acquisition unit changes the acquired distance information based on the reliability information.

Abstract: Methods and systems for luggage visualization perform virtual unpacking by visually moving an object image away from its original pose. A scanned 3D volume is segmented guided by a confidence measure to create a label volume whose voxels specify the detected object IDs. The luggage dataset and the label volume are visualized by volume rendering. Using an automatic coloring algorithm, any pair of objects whose projections are adjacent in an image are assigned distinct hues. A layered framework efficiently renders a scene mixed with packed luggage, animated unpacking objects, and already unpacked objects put aside for further inspection. A GPU is used to automatically select objects that are not blocked by others and can be unpacked.

Abstract: A detection control device includes a detection target image acquiring unit that acquires an imaged image resulting from imaging a specimen that is to be subjected to the detection of the target cell, a processing target region setting unit that sets a plurality of processing target regions in the imaged image, a processing target region distributing unit that distributes, in accordance with a distribution amount of data to each of a plurality of processing devices that detect the target cell in an image region, the plurality of processing target regions to the plurality of processing devices, and a detection result acquiring unit that acquires, from each of the plurality of processing devices, a result of a process of detecting the target cell in the processing target regions distributed to the respective plurality of processing devices.

Abstract: Disclosed is a learning device. A feature-quantity calculation unit extracts a feature quantity from each feature point of a learning image. An acquisition unit acquires a classifier already obtained by learning as a transfer classifier. A classifier generation unit substitutes feature quantities into weak classifiers constituting the transfer classifier, calculates error rates of the weak classifiers on the basis of classification results of the weak classifiers and a weight of the learning image, and iterates a process of selecting a weak classifier of which the error rate is minimized a plurality of times. In addition, the classifier generation unit generates a classifier for detecting a detection target by linearly coupling a plurality of selected weak classifiers.

Abstract: Website structure creation. A site navigation structure sketch is created by a free-form drawing using a drawing tool being executed on a computing device. The site navigation structure sketch is stored in pixel form, and processed by transforming the site navigation structure sketch by vectorizing the sketch into a predefined structured representation. A site navigation structure is created based on the predefined structured representation, and the site navigation structure is integrated into a web portal site structure.

Abstract: Systems and methods are described for receiving a current image that partially overlaps with a previous image of a series of images of an original document; performing optical character recognition (OCR) of the current image, producing an OCR text and a corresponding text layout; identifying textual artifacts in the current and previous images, each represented by a sequence of symbols having a frequency of occurrence within the OCR text below a threshold frequency; identifying corresponding base points associated with textual artifacts; identifying parameters of a coordinate transformation converting coordinates of the previous image into coordinates of the current image; associating part of the OCR text with a cluster of symbol sequences, wherein the symbol sequences are produced by processing previously received images; identifying an order of clusters of symbol sequences reflecting a layout of the original document; and producing a resulting OCR text representing a portion of the original document.

Abstract: Methods and apparatus relating to enabling augmented reality applications using eye gaze tracking are disclosed. An exemplary method according to the disclosure includes displaying an image to a user of a scene viewable by the user, receiving information indicative of an eye gaze of the user, determining an area of interest within the image based on the eye gaze information, determining an image segment based on the area of interest, initiating an object recognition process on the image segment, and displaying results of the object recognition process.

Abstract: An image processing device that converts original image data to target image data is provided. The image processing device includes: a static random access memory (SRAM); an image scaling circuit that generates scaled image data according to the original image data and stores the scaled image data to the SRAM; and a video encoding circuit that accesses the scaled image data from the SRAM and encodes the accessed scaled image data to generate the target image data. The target image data corresponds to an image frame. A part of the target image data is intra frame data encoded by an intra frame compression method, and the other part of the target image data is predicted frame data encoded by a predicted frame compression method.

Abstract: A method of segmenting a target organ of the body in medical images, the method comprising: providing a probabilistic atlas with probabilities for a presence of the target organ at different locations relative to a bounding region of the target organ; providing one or more medical test images; for each test image, identifying, at least provisionally, a location of a bounding region for the target organ in the test image; and for each test image, using the probabilities from the probabilistic atlas to segment the target organ in the test image, with the segmenting performed by a data processor.

Abstract: Method and apparatus for generating superpixel clusters A method and an apparatus (20) for generating a superpixel cluster for an image are described. A clustering unit (23) generates (10) an initial superpixel cluster based on an initial cluster center. The clustering unit (23) further generates (12) one or more further superpixel clusters based on one or more further cluster centers. A determining unit (24) determines (11) the one or more further cluster centers from the initial superpixel cluster. A merger (25) then merges (13) the initial superpixel cluster and the one or more further superpixel clusters to obtain the superpixel cluster for the image.

Abstract: Certain embodiments provide a computer system for determining a registration mapping between a novel medical image and a reference medical image, the computer system comprising: a storage system adapted to store data representing the novel medical image and the reference medical image and variance data for a plurality of different locations in the reference medical image representing a statistical variation for corresponding locations identified in a plurality of training medical images; and a processor unit operable to execute machine readable instructions to determine a registration mapping between the novel medical image and the reference medical image in a manner that takes account of the variance data for the plurality of different locations in the reference medical image.

Abstract: A computerized method for annotating at least one feature of an image of a view, includes the steps of obtaining the image with an image sensor of a portable device, and retrieving at least one condition. Based on the at least one condition, the method automatically selects a feature identification method among a plurality of features identification methods. It then applies the feature identification method for identifying the at least one feature, and annotates some of the identified features.

Abstract: A system and method for providing image guidance for planning approach paths for one or more medical devices at a target location. The system can receive a volumetric medical image, determine a density of content within the volumetric medical image, receive an indication of a target location within the volumetric medical image, identify obstructing objects within the volumetric medical image, and determine a plurality of pathways from an approach region of the volumetric medical image to the target location. The system can cause a display to concurrently display the volumetric medical image and a plurality of emplacements for one or more medical imaging devices.

Abstract: A first set of attributes (e.g., style) is generated through pre-trained single column neural networks and leveraged to regularize the training process of a regularized double-column convolutional neural network (RDCNN). Parameters of the first column (e.g., style) of the RDCNN are fixed during RDCNN training. Parameters of the second column (e.g., aesthetics) are fine-tuned while training the RDCNN and the learning process is supervised by the label identified by the second column (e.g., aesthetics). Thus, features of the images may be leveraged to boost classification accuracy of other features by learning a RDCNN.

Abstract: A paragraph alignment detection engine and a section reconstruction engine. The paragraph alignment detection engine determines the paragraph alignment of a paragraph and updates the paragraph alignment property of the paragraph in the data store for single line and multi-line paragraphs. The paragraph alignment detection engine employs per paragraph comparisons and relative comparisons to other paragraphs to determine the paragraph alignment of a single line paragraph. The paragraph alignment detection engine employs per paragraph comparisons and relative comparisons of the lines of a paragraph to determine the paragraph alignment of a multi-line paragraph. The section reconstruction engine minimizes the number of sections created in the flow format document by identifying the columns on each page, combining contiguous pages with the same column layout into a single section, and creating alternative objects to contain regions associated special cases in lieu of creating additional sections.

Abstract: A camera system and method provide a trailing motion blur effect without the use of a flash. In one embodiment, a variable light-transmissive filter is positioned in front of an image sensor. While the light-transmissive property of the filter is lower, the blurred image of the object in motion is produced. The light-transmissive property of the filter is increased during the exposure in order to produce a clearer image of the object. In a method, a plurality of images taken in rapid succession are combined to produce a clear image of the object with a trailing blurred image of the object.

Abstract: According to some embodiments, system and methods are provided comprising determining one or more image locations at which motion occurred within an image volume of an object containing movable anatomical features prior to segmenting the movable anatomical features; estimating motion at each of the one or more image locations; correcting one or more motion artifacts in the image volume at each of the one or more image locations, where motion was estimated resulting in the generation of a final motion compensated image; and segmenting one or more features of interest in the final motion compensated image. Numerous other aspects are provided.

Abstract: A method for processing a picture is disclosed. The method according to the invention comprises: segmenting the picture into a background layer and at least another layer; selecting a background template of color harmony for the background layer; determining one template of color harmony for each of the at least another layer from the background template of color harmony depending on the distance between the background layer and said at least another layer. processing the picture by mapping the colors of each segmented layer using the corresponding template of color harmony.

Abstract: Based on a multi-step process, an automatic hair segmentation method and system are disclosed. By extracting various information components from an image, including background color, face position, hair color, skin color, and skin mask, a heuristic-based method is created for the detection and segmentation of hair that can detect hair with an accuracy of approximately 75% and with a false hair overestimation error of 34%. Furthermore, it is shown that down sampling the image down to a face width of 25 px results in a 73% reduction in computation time with insignificant change in detection accuracy.

Abstract: A field extraction template simplifies the creation of field extraction rules by providing a user with a set of field names commonly assigned to a certain type of data, as well as guidance on how to extract values for those fields. These field extraction rules, in turn, facilitate access to certain “chunks” of the data, or to information derived from those chunks, through named fields. A field extraction template comprises at least a set of field names and ordering data for the field names. The ordering data indicates index positions that are associated with at least some of the field names. A delimiter is specified for splitting data items into arrays of chunks. The chunk of a data item that belongs to a given field name is the chunk whose position within the item's array of chunks is equivalent to the index position associated with the given field name.

Abstract: An image display apparatus for displaying an image containing a plurality of objects includes a setting unit configured to set a display magnification and a display position according to an attribute of a display target object when a first display mode for displaying each object included in the image is specified, and a display control unit configured to perform control to display on a screen the image containing the display target object based on the display magnification and the display position set by the setting unit.

Abstract: A method of updating a scene model for a foreground segmentation of an input image captured from a camera, is disclosed. One or more visual elements of the input image are determined. A spatial relationship between at least one of the visual elements and the scene model for a foreground segmentation of the input image is determined. The method updates the scene model for determining the foreground segmentation of the input image based on the determined spatial relationship.

Abstract: A system for determining the prognosis of a patient suffering from pulmonary embolism is provided. The system may include at least one computer system configure to receive patient specific data regarding his pulmonary embolism status. The at least one computer system may be further configured to create a model of the patient's heart, with at least information of the two ventricles, and to determine the ratio of sizes of the ventricles. The system will then report such ratio to the clinician or report a risk index of clinical outcome for such patient.

Abstract: There is described herein an image segmentation technique using an iterative process. A contour, which begins with a single point that expands into a hollow shape, is iteratively deformed into a defined structure. As the contour is deformed, various constraints are applied to points along the contour to dictate its rate of change and direction of change are modified dynamically. The constraints may be modified after one or more iterations, at each point along the contour, in accordance with newly measured or determined data.

Abstract: First, the position and the size of a target photographic subject such as the face of a person in an original image acquired by image capturing is acquired, and then mask information indicating a relation between a position to be subjected to predetermined image processing in the original image and the strength of the predetermined image processing is generated based on the acquired position and size. Subsequently, a new image is generated by predetermined image processing of a different strength being performed on each position in the original image by use of the generated mask information. In the generation of the new image, for example, one of the original image and a processed image acquired by the predetermined image processing being performed on the original image is combined with the other image by use of transmissivity indicated by the mask information for each pixel.

Abstract: A method for image processing includes obtaining a mask of a stroke from an image and identifying a plurality of cross edges for the stroke based on the mask and a reference line. The plurality of cross edges includes a group of adjacent cross edges that intersect the reference line. The method further includes (a) calculating a first vector based on positions of at least two of the cross edges in the group, (b) expanding the group, based on the first vector, to include cross edges adjacent to the group that do not intersect the reference line, (c) calculating a second vector based on positions of at least two of the cross edges in the expanded group, and (d) expanding the expanded group, based on the second vector, to include a second group of adjacent cross edges nearby the expanded group that do not intersect the reference line.

Abstract: The present disclosure provides virtual replacement of the texture of an apparel with a different texture, also taking care of the wrinkles, body shape etc. The apparel worn by the user is identified/localized in all the frames of a video file, and replaced with a pre-identified texture from the catalogue. Multiple histograms are calculated to localize the region of the apparel using Correlation coefficient. The variation in light intensity is used to capture the wrinkles and folds etc on the apparel.

Abstract: A document processor for associating mentioned items with each other in two documents. A conversion unit converts at least a portion of mentioned items expressed in table format among the mentioned items included in the two documents. An association processing unit associates mentioned items with each other in the two documents including at least some mentioned items converted to text format. The conversion unit converts each of a plurality of cells included in the mentioned items expressed in table format to separate mentioned items in text format.

Abstract: A medical imaging apparatus and a method of processing a medical image. The medical imaging apparatus includes: an image processor configured to respectively determine color values for subregions of an organ based on at least two of first through third analysis values for each of the subregions. A display outputs display of a first map image obtained by respectively mapping the determined color values to the subregions and indicating the determined color values on the subregions. In the case where the image of the organ is a heart, the different parts of the heart correspond to respective subregions, and an indication of a probability of disease corresponds to a plurality of colors in a color map of the regions.

Abstract: System and method for active contour segmentation where an image for a structure and an initial position on the image are received, a multi-scale image representation including successive image levels each having associated therewith a representation of the image is computed, a representation of the image at a given image level having a different image resolution that that of a representation of the image at a subsequent level, a given one of the levels at which noise in the image is removed is identified, the initial position is set as a current contour and the given level as a current level, the current contour is deformed at the current level to expand into an expanded contour matching a shape of the structure, the expanded contour is set as the current contour and the subsequent level as the current level, and the steps are repeated until the last level is reached.

Abstract: Systems and methods are provided to improve map type data visualizations in which custom images are used as background against which datasets are visualized. In various examples, coordinate systems are automatically generated and applied to the custom image based on the positional data of the dataset. The custom image may replace a pre-existing background, and the plotting of the dataset may be automatically modified based on differences between the pre-existing background and the custom image. Aspects provide for enhanced user control of the map type visualization to further modify how datasets are plotted and how visualizations are presented. Example modifications include offsets and scales, which may be automatically determined and applied or set manually. In some examples, the custom image may be part of a set of image tiles, which are operable to provide various levels of detail from several viewing angles and heights.

Abstract: The present disclosure relates to a video monitoring method, a video monitoring system and a computer program product. The video monitoring method comprises: obtaining video data collected by a video data collecting apparatus; and based on pre-set scene information and the video data, performing determination and statistics of monitored objects in a scene to be monitored corresponding to the scene information.

Abstract: A method and apparatus is provided to denoise material-decomposition data generated using projection data from a spectral computed tomography scanner. A whitening transform is used to transform the material-decomposition data into uncorrelated components and perform denoising on the uncorrelated components. Using different denoising parameters for the various uncorrelated components, a flattening can be achieved for the standard deviation of the noise as a function of X-ray energy, which can be determined using mono-energetic images derived from the material-decomposition data. The whitening transformation and the denoising can be applied the material-decomposition sinograms and/or to material-decomposition images reconstructed from the material-decomposition sinograms.

Abstract: Techniques of detecting text in video are disclosed. In some embodiments, a portion of video content can be identified as having text. Text within the identified portion of the video content can be identified. A category for the identified text can be determined. In some embodiments, a determination is made as to whether the video content satisfies at least one predetermined condition, and the portion of video content is identified as having text in response to a determination that the video content satisfies the predetermined condition(s). In some embodiments, the predetermined condition(s) comprises at least one of a minimum level of clarity, a minimum level of contrast, and a minimum level of content stability across multiple frames. In some embodiments, additional information corresponding to the video content is determined based on the identified text and the determined category.

Abstract: A first image is received. An initial label is assigned to at least some pixels in the first image, including by assigning a first label to a first pixel. A determination is made, based at least in part on labels of spectral neighbors of the first pixel, that the first pixel's label should be replaced with a different label. The first pixel's label is updated with the different label.