Abstract: Aspects of the present disclosure provide techniques for document classification through signal processing. Embodiments include receiving a document for classification. Embodiments include generating an image of the document. Embodiments include producing a signal representation of the document based on numbers of non-white pixels in each horizontal scan line or vertical scan line of the image of the document. Embodiments include comparing the signal representation of the document to signal representations of previously-classified documents. Embodiments include determining, based on the comparing, a classification for the document. Embodiments include performing additional processing with respect to the document based on the classification for the document.

Abstract: A three-dimensional data encoding method includes: encoding first information of a first current node or second information of a second current node, the first current node being included in an N-ary tree structure of first three-dimensional points included in a first three-dimensional point cloud, N being 2 or 4, the second current node being included in an octree structure of second three-dimensional points included in a second three-dimensional point cloud. The encoding of the first information or the second information includes encoding the first information using a first encoding pattern including a pattern common to a second encoding pattern used in encoding the second information.

Abstract: Items are identified in a waste stream for purposes of recycling, using deterministic and/or probabilistic techniques. Imagery of the waste stream from multiple viewpoints permit creation of a 3D depth draped image representation, from which one or more 2D planes can be synthesized. Phase-coherent patches of recoverable encoded data can be identified from among soiled and crumpled object surfaces, and used in combination to recover object identification information. Recognition of certain items can trigger further image processing that is specific to such items. (Detection of a catsup bottle, for example, can trigger image analysis to discern the presence of catsup residue.) Information about recognized objects can be provided to external data customers, e.g., to track grey market diversion of particular products into unlicensed territories. These and other features and advantages, which can be used alone or in combination, are detailed herein.

Abstract: A method of obtaining a tree-based point cloud compression (PCC) media stream using moving picture experts group (MPEG)-dynamic adaptive streaming over http (DASH), is performed by at least one processor and includes obtaining, from a server, a media representation description (MPD) comprising a PCC uniform resource indicator (URI) and a PCC three-dimensional (3D) descriptor, and determining whether the PCC URI comprised in the obtained MPD indicates that the tree-based PCC media stream is to be accessed. The method further includes, based on the PCC URI being determined to indicate that the tree-based PCC media stream is to be accessed, obtaining, from the server, the tree-based PCC media stream, using the PCC 3D descriptor comprised in the obtained MPD.

Abstract: A method for characterization of coronary plaque tissue data and perivascular tissue data using image data gathered from a computed tomography (CT) scan along a blood vessel, the image information including radiodensity values of coronary plaque and perivascular tissue located adjacent to the coronary plaque, the method comprising quantifying radiodensity in regions of coronary plaque, quantifying, radiodensity in at least one region of corresponding perivascular tissue adjacent to the coronary plaque, determining gradients of the quantified radiodensity values within the coronary plaque and the quantified radiodensity values within the corresponding perivascular tissue, and determining a ratio of the quantified radiodensity values within the coronary plaque and the corresponding perivascular tissue; and characterizing the coronary plaque by analyzing a gradient of the quantified radiodensity values in the coronary plaque and the corresponding perivascular, and/or the ratio of the coronary plaque radiodensity

Abstract: A three-dimensional data encoding method includes: obtaining third point cloud data that is a combination of first point cloud data and second point cloud data, and includes geometry information of each of three-dimensional points included in the third point cloud data, and identification information indicating which of the first point cloud data and the second point cloud data each of the three-dimensional points belongs to; and generating encoded data by encoding the third point cloud data obtained. In the generating, identification information of each of the three-dimensional points is encoded as attribute information.

Abstract: At least one embodiment relates to a method comprising encoding a projection mode associated with an image region of a first depth image, said projection mode indicating if said image region stores either the minimum or the maximum depth value of at least one orthogonally projected 3D point of the point cloud along a same projection direction; and encoding said depth minimum and maximum depth values as a function of said projection mode.

Abstract: A method and apparatus for coding information of a point cloud may be performed by at least one processor and comprises: obtaining the point cloud including a set of points in a three-dimensional space; partitioning the point cloud into a tree structure comprising a plurality of nodes at different depths; encoding geometry information of the nodes; and encoding attribute information of the nodes before the entire point cloud is partitioned.

Abstract: Systems and methods for automated resolution of over-specification and under-specification in a knowledge graph are disclosed. In embodiments, a method includes: determining, by a computing device, that a size of an object cluster of a knowledge graph meets a threshold value indicating under-specification of a knowledge base of the knowledge graph; determining, by the computing device, sub-classes for objects of the knowledge graph; re-initializing, by the computing device, the knowledge graph based on the sub-classes to generate a refined knowledge graph, wherein the size of the object cluster is reduced in the refined knowledge graph; and generating, by the computing device, an output based on information determined from the refined knowledge graph.

Abstract: Disclosed is a skeleton-based dynamic point cloud estimation system for sequence compression, capable of generating a three-dimensional skeleton and a motion vector thereof from a group of point cloud frames to estimate a point cloud of a next frame, and compressing a point cloud sequence by obtaining a residual between the estimated point cloud and an original point cloud. Since the point cloud is estimated by using the motion vector of the skeleton, and the sequence is compressed by using a difference from an original value as the residual, an amount of point clouds to be input to a compressor is significantly reduced, so that compression efficiency is improved.

Abstract: A method and apparatus for vascular assessment are disclosed. The apparatus, in some embodiments, receives a plurality of 2-D angiographic images of a portion of a vasculature of a subject, and processes the images to produce a stenotic model over the vasculature. The stenotic model has measurements of the vasculature at one or more locations along vessels of the vasculature. The apparatus, in some embodiments, determines a flow characteristic of the stenotic model and calculates an index indicative of vascular function, based, at least in part, on the flow characteristic in the stenotic model.

Abstract: A method for generating and evaluating N-to-1 mappings between spatial point sets in nD, n=2 or 3, implemented on a computing device comprising a programmable general purpose processor and a programmable data-parallel coprocessor and a memory coupled with them.

Abstract: An analyzing apparatus according to an embodiment includes processing circuitry. The processing circuitry is configured to calculate a tissue characteristic parameter value with respect to each of a plurality of positions within a region of interest, by analyzing a result of a scan performed on a patient. The processing circuitry is configured to determine a measurement region in the region of interest by performing an analysis while using the tissue characteristic parameter values. The processing circuitry is configured to calculate a statistic value of the tissue characteristic parameter values in the measurement region.

Abstract: Systems and methods that yield highly-accurate classification of acoustic and other non-image events, involving pre-processing data from one or more transducers and generating a visual representation of the source as well as associated features and processing, are disclosed. According to certain exemplary implementations herein, such pre-processing steps may be utilized in situations where 1) all impulsive acoustic events have many features in common due to their point source origin and impulsive nature, and/or 2) the error rates that are considered acceptable in general purpose image classification are much higher than the acceptable levels in automatic impulsive incident classification. Further, according to some aspects, the data may be pre-processed in various ways, such as to remove extraneous or irrelevant details and/or perform any required rotation, alignment, scaling, etc. tasks, such that these tasks do not need to be “learned” in a less direct and more expensive manner in the neural network.

Abstract: A system and method for tracking and annotating products in a store shelf is disclosed. The method includes receiving a signal feed from a group of sensors including at least one depth sensor and a sensor capable of measuring distance, determining an area of interest in the shelf, determining a polygonal representation of at least one product present in the area of interest, and assigning a unique polygonal profile to at least one product, and/or storing the unique profile in a database. Such The unique polygonal profile may include a unique ID obtained from a group of attributes. The unique profile may include at least a unique position and a unique centroid. The group of sensors may further include video cameras, thermal, and chemical sensors.

Abstract: A method and system for communicating estimated blood loss parameters of a patient to a user, the method comprising: receiving data representative of an image, of a fluid receiver; automatically detecting a region within the image associated with a volume of fluid received at the fluid receiver, the volume of fluid including a blood component; calculating an estimated amount of the blood component present in the volume of fluid based upon a color parameter represented in the region, and determining a bias error associated with the estimated amount of the blood component; updating an analysis of an aggregate amount of the blood component and an aggregate bias error associated with blood loss of the patient, based upon the estimated amount of the blood component and the bias error; and providing information from the analysis of the aggregate amount of the blood component and the aggregate bias error, to the user.

Abstract: Systems and methods for detecting eye information are described. According to one embodiment, a detecting device includes a determination unit, a generation unit, and a detection unit. The determination unit is configured to determine whether a user wears glasses from an input image by using a first neural network including a plurality of filters. The generation unit is configured to generate a face image in which the glasses are removed from the input image by using a learned generator when a glasses-wearing signal is received from the determination unit. The detection unit is configured to generate an eye image by receiving the input image from the determination unit or receiving the face image from the generation unit and detect a pupil from the eye image using a second neural network including a plurality of filters.

Abstract: Bordering pixels delineating a texture hole region are identified in a target image. Depth values of the bordering pixels are automatically clustered into two depth value clusters. A specific estimation direction is selected from multiple candidate estimation directions for a texture hole pixel in a texture hole region. A depth value of the texture hole pixel is estimated by interpolating depth values of two bordering background pixels in the specific estimation direction. The estimated depth value is used to warp the texture hole pixel into a reference view represented by a temporal reference image. A pixel value of the texture hole pixel is predicted based on a reference pixel value of a reference pixel from the reference image to which the texture hole pixel is warped using the estimated depth value.

Abstract: Multiple single lens distortion regions and one or more boundary regions may be determined within an image. Individual single lens distortion regions may have pixel displacement defined by a single lens distortion and individual boundary regions may have pixel displacement defined by a blend of at least two lens distortions. Multiple lens distortions may be simulated within the image by shifting pixels of the image input positions to output positions based on locations of pixels within a single lens distortion region and the corresponding single lens distortion or within a boundary region and a blend of corresponding lens distortions.

Abstract: Embodiments of the present disclosure disclose a method and apparatus for identifying a video. A specific embodiment of the method includes: acquiring a predetermined number of video frames from a video to be identified to obtain a video frame sequence; performing the following processing step: importing the video frame sequence into a pre-trained video identification model to obtain a classification tag probability corresponding to the video frame sequence, wherein the classification tag probability is used to characterize a probability of identifying a corresponding tag category of the video to be identified; and setting, in response to the classification tag probability being greater than or equal to a preset identification accuracy threshold, a video tag for the video to be identified according to the classification tag probability, or else increasing the number of video frames in the video frame sequence and continuing to perform the above processing step.