Abstract: A three-dimensional data encoding method includes: determining whether a first valid node count is greater than or equal to a first threshold value predetermined, the first valid node count being a total number of valid nodes that are nodes each including a three-dimensional point, the valid nodes being included in first nodes belonging to a layer higher than a layer of a current node in an N-ary tree structure of three-dimensional points included in point cloud data, N being an integer greater than or equal to 2; and, when the first valid node count is greater than or equal to the first threshold value, performing first encoding on attribute information of the current node, the first encoding including a prediction process in which second nodes are used, the second nodes including a parent node of the current node and belonging to a same layer as the parent node.

Abstract: During enrollment to a biometric identification system an image of at least a portion of a user is acquired and processed to determine a compact representation (CR). The CR may be stored for later comparison to identify the user, to train embedding models, and so forth. The image is not stored. A machine learning system that comprises a variational autoencoder is trained to produce the CR with a loss function that includes a distortion loss, an embedding distance loss, and in some implementations a bitrate loss. The trained encoder is used to determine the CR from an input image, while the decoder is not stored. The CR contains sufficient information to be used as training data for embedding models, instead of acquired images. Training may be more computationally efficient using the CR. Recognition comparisons may be more efficiently performed using the CR, compared to query images.

Abstract: An electronic apparatus may include a memory, and a processor configured to perform an operation between a first index data and a second index data corresponding to a first data set and a second data set, respectively, to acquire first output data; identify a sparsity ratio of the first output data; based on the sparsity ratio exceeding the threshold value, determine a first method of processing the first data set and the second data set using the first data set and the second data set, as an operation method of the electronic apparatus; and based on the sparsity ratio being equal to or less than the threshold value, identify valid data included in the first index data and the second index data, and determine a second method of processing the first data set and the second data set using the valid data, as the operation method of the electronic apparatus.

Abstract: A method of video decoding comprises receiving a bitstream including a SPS, a PPS, and subpictures associated with a subpicture ID mapping. A determination is made of whether an SPS flag has a first or second value. The SPS flag having the first value specifies that the subpicture ID mapping is signaled in the SPS, and the second value specifies that the subpicture ID mapping is signaled in the PPS. The subpicture ID mapping is obtained from the SPS when the SPS flag has the first value and from the PPS when the SPS flag has the second value or the PPS flag has the first value.

Abstract: Both of a high compression ratio and a high processing speed are achievable. In a data compression and decompression system that includes a parallel processing device performing a plurality of processes in parallel, the parallel processing device divides original data into a plurality of data by a predetermined unit. The parallel processing device performs coding processes on the plurality of data in parallel using a predetermined model to create a plurality of coded data. The parallel processing device creates compressed data of the original data from the plurality of coded data.

Abstract: Content-adaptive online training for end-to-end (E2E) neural image compression (NIC) using a neural network performed by at least one processor, is provided, including receiving an input image, to an E2E NIC framework, including one or more blocks, preprocessing a first neural network of the E2E NIC framework, based on the one or more blocks, computing updated parameters using the preprocessed first neural network, encoding the one or more blocks and the updated parameters, updating the first neural network based on the encoded updated parameters, and generating a compressed representation of the encoded one or more blocks using the updated first neural network.

Abstract: Methods and apparatus for compressing data streams. In an embodiment, a method includes calculating a decomposition of matrix data to generate eigenvectors and associated eigenvalues, determining clusters of the eigenvectors based on weighting the eigenvalues, calculating an eigenvector centroid for each cluster so that a dictionary of centroids is generated, and tagging the eigenvectors with tags, respectively, that identify an associated eigenvector centroid for each eigenvector.

Abstract: Method for correcting amplitude underestimations in time series, which were measured on sequences of multi-dimensional signals, wherein the multi-dimensional signals were sampled with one or more integrating sensors, such that the time series are degraded by a center of gravity/centroid formation, which can be characterized by a duty cycle of the integrating sensors. According to the invention, a centroid formation of the time series is corrected based on the duty cycle, the duty cycle or the integration time of the sensors.

Abstract: A method, apparatus, article of manufacture, and a memory structure for signaling extension functions used in decoding a sequence comprising a plurality of pictures, each picture processed at least in part according to a picture parameter set is disclosed. In one embodiment, the method comprises reading a first extension flag signaling a first extension function in the processing of the sequence and determining if the first extension flag has a first value. Further, the method reads a second extension flag signaling a second extension function in the processing of the sequence and performs the second extension function according to the read second extension flag only if the first extension flag has a first value.

Abstract: Disclosed herein is a method for compressing point cloud data. The method includes quantizing input point cloud data, generating a global motion matrix based on the quantized point cloud data, applying the global motion matrix based on whether local motion compression is performed, and compressing data to which the global motion matrix is applied.

Abstract: A point cloud data transmission method, according to embodiments, may comprise the steps of: encoding point cloud data; and transmitting the point cloud data. The present invention, according to embodiments, may comprise the steps of: receiving point cloud data; and decoding the point cloud data.

Abstract: Technologies for conversion and transmission of data files from a mainframe computing system for big data ingestion. Secure electronic payment transactional data is extracted in a mainframe format, such as Extended Binary Coded Decimal Interchange Code (EBCDIC). The electronic payment transactional data is converted to American Standard Code for Information Interchange (ASCII) code prior to transmission to a big data analytics computing system. The converted data is encrypting and transmitted for ingestion into a big data analytics computing system.

Abstract: The present application relates to detecting if video images captured by a camera are depicting a live scene or a recorded video played on a monitor, display or computer screen, which is setup to hide the scene from the camera. Metadata regarding the mapping operation used to transform image data between different intensity ranges, or bit depths, is included with the video and evaluated in order to determine if a video replay attack has taken place.

Abstract: An image is split into a plurality of blocks of various sizes and a subdivision level counter is associated to each of the blocks. The value of this subdivision level counter for a block is representative of the size of the block and is used to determine the quantization parameter for the block. The value is propagated for each subdivision and incremented according the type of the subdivision. When the image is split, an analysis is done according to the subdivision level counter, a maximal value of subdivision and the type of split, in order to determine the start of a new quantization group and when it is the case, the current position of the partition is propagated to the further split partitions to be stored with these partition and serve in the prediction process when decoding.

Abstract: Techniques are described for encoding a normal map of a computer graphics texture. identify a normal map associated with a computer graphics texture. X and Z components of the normal map are associated with respective first and second channels with each channel being represented as two respective pair of endpoint values and 6-bits of interpolation indices. The channels including negative Z values are projected to octahedral space with interpolated index values being allowed outside the range of [?1, 1], and a symmetry operation executed to bring the interpolation indices values into the range [?1, 1].

Abstract: A method for generating at least one segment of a video by an electronic device is provided. The method includes identifying at least one of a context associated with the video and an interaction of a user in connection with the video, analyzing at least one parameter in at least one frame of the video with reference to at least one of the context and the interaction of the user, wherein the at least one parameter includes at least one of a subject, an environment, an action of the subject, and an object, determining the at least one frame in which a change in the at least one parameter occurs, and generating at least one segment of the video comprising the at least one frame in which the parameter changed as a temporal boundary of the at least one segment.

Abstract: The present disclosure generally relates to the field of camera surveillance, and in particular to a method and control unit for controlling a bitrate of a video stream captured with an image acquisition device.

Abstract: Disclosed herein is a method of transmitting point cloud data. The method may include encoding geometry data of the point cloud data, encoding attribute data of the point cloud data based on the geometry data, and transmitting the encoded geometry data, the encoded attribute data and signaling data, the geometry encoding includes splitting the geometry data into one or more prediction units, and inter-prediction encoding the geometry data by selectively applying a motion vector to each of the split prediction units, and the signaling data includes information for identifying whether the motion vector is applied for each prediction unit.

Abstract: Systems and methods are provided for distributed video storage and search over edge computing devices having a short-term memory and a long-term memory. The method may comprise caching a first portion of data on a first device. The method may further comprise determining, at a second device, whether the first device has the first portion of data. The determining may be based on whether the first piece of data satisfies a specified criterion. The method may further comprise sending the data, or a portion of the data, and/or a representation of the data from the first device to a third device.

Abstract: A method for controlling a display apparatus, including calculating at least one of a complexity or an average picture level (APL) of the received image signal, determining a final shift time of pixels based on the at least one of the calculated complexity or APL of the image signal, and shifting the received image signal in a horizontal direction or a vertical direction and outputting the shifted image signal, based on track coordinate information stored in a memory and the determined final shift time of the pixels.

Abstract: Methods and devices are provided to encode and decode a data stream carrying data representative of a three-dimensional scene, the data stream comprising color pictures packed in a color image; depth pictures packed in a depth image; and a set of patch data items comprising de-projection data; data for retrieving a color picture in the color image and geometry data. Two types of geometry data are possible. The first type of data describes how to retrieve a depth picture in the depth image. The second type of data comprises an identifier of a parametric function and a list of parameter values for the identified parametric function.

Abstract: A method for non-destructive ripeness identification of kiwifruit based on machine vision learning may include: collecting kiwifruit data to obtain an original data set by collecting images of 40-80 kiwifruits in the same period of time over 3-6 days, recording a label, which comprises ripeness information for each of the images, and saving each of the images with the label; extracting the color and the texture of a kiwifruit skin from each of the images in the original data set; and training a deep learning model to learn a connection between the color and the texture of the kiwifruit skin and the ripeness information of the corresponding kiwifruit using the color and the texture of the kiwifruit skin extracted from each of the images and the label.

Abstract: Systems and methods for deferring post-process effects in video encoding are disclosed. The systems and methods are capable of calculating the capability of client hardware to defer load, and summing a known load of one or more deferral candidates to evaluate how many post-process deferral candidates are capable of being deferred to client hardware. The systems and methods are also capable of sending an updated deferral list of post-processes to a remote server that can apply the list of deferred post-process candidates during the post-processing phase of a first video frame.

Abstract: At least one embodiment relates to a method comprising obtaining an occupancy information indicating if blocks of a 2D regular grid defined over a projection plane are associated with depth information of an orthogonal projection of at least one point of a point cloud onto said projection plane; embedding said occupancy information in an image storing said depth information; and encoding said image.

Abstract: An image encoding/decoding method and apparatus are provided. An image decoding method may comprise obtaining, from a bitstream, a feature set extracted from an input image using an artificial neural network-based feature extraction method, first information on importance of each of a plurality of channels included in the feature set and second information on a quantization method of the feature set, decoding the feature set, the first information and the second information, and dequantizing the decoded feature set based on the decoded first information and the decoded second information. The second information may comprise the number of quantization bits of each of the plurality of channels, and the number of quantization bits may be determined based on the importance of each of the plurality of channels.

Abstract: A visual perception system includes a scanning camera, color sensor with color filter array (CFA), and a classifier node. The camera captures full color pixel images of a target object, e.g., a traffic light, and processes the pixel images through a narrow band pass filter (BPF), such that the narrow BPF outputs monochromatic images of the target object. The color sensor and CFA receive the monochromatic images. The color sensor has at least three color channels each corresponding to different colors of spectral data in the monochromatic images. The classifier node uses a predetermined classification decision tree to classify constituent pixels of the monochromatic images into different color bins as a corresponding color of interest. The color of interest may be used to perform a control action, e.g., via an automated driver assist system (ADAS) control unit or an indicator device.

Abstract: A system and method for data compression with protocol adaptation, that utilizes a codebook generator which leverages one or more machine/deep learning algorithms trained on at least a plurality of protocol policies in order to generate a protocol appendix and codebook, wherein original data is encoded by an encoder according to the codebook and sent to a decoder, but instead of just decoding the data according to the codebook to reconstruct the original data, data manipulation rules such as mapping and transformation are applied at the decoding stage to transform the decoded data into protocol formatted data.

Abstract: A system and method for compressing and decompressing image data, which provides better compression and minimal representation of the input image with minimal loss compared to previous Deep Learning codecs. The system can provide for quantization during training, flexible addition of filters, and conditional complexity of image compression. The Deep Learning codec generates codes that are directly usable with Machine Learning algorithms, thus boosting the performance of Machine Learning algorithms.

Abstract: Interpolation logic described herein provides a good approximation to a bicubic interpolation, which is generally smoother than bilinear interpolation, without performing all the calculations normally needed for a bicubic interpolation. This allows an approximation of smooth bicubic interpolation to be performed on devices (e.g. mobile devices) which have limited processing resources. At each of a set of predetermined interpolation positions within an array of data points, a set of predetermined weights represent a bicubic interpolation which can be applied to the data points. For a plurality of the predetermined interpolation positions which surround the sampling position, the corresponding sets of predetermined weights and the data points are used to determine a plurality of surrounding interpolated values which represent results of performing the bicubic interpolation at the surrounding predetermined interpolation positions.

Abstract: An encoder includes circuitry and a memory coupled to the circuitry. The circuitry, in operation, determines whether or not a ternary split process of splitting a block into three sub blocks in a first direction parallel to a first longer side of the block is allowed by comparing a size of a second shorter side of the block to a minimum threshold value. The circuitry, responsive to the ternary split process being allowed, writes, into a bitstream, a split direction parameter indicative of a splitting direction. The circuitry, in operation, splits the block into a plurality of sub blocks in a direction indicated by the split direction parameter; and encodes the plurality of sub blocks. The minimum threshold value corresponds to a minimum size supported in a transform process.

Abstract: A method may include obtaining an image of a subject and determining a plurality of image blocks in the image. The method may also include extracting grayscale features from each of the plurality of image blocks and determining, based on the grayscale features, a segmentation threshold. The method may further include segmenting, based on the segmentation threshold, the image.

Abstract: Methods and devices for coding point cloud data using volume trees and predicted-point trees. In one embodiment of the disclosure, a method of encoding a point cloud data to generate a bitstream of compressed point cloud data representing a three-dimensional location of a physical object is provided, the point cloud data being located within a volumetric space. The method includes compressing a first part of the point cloud data represented by a first tree of a first type; determining for a given node of the first tree if an assignation to a second type of tree is enabled, said given node still being processed for the first tree; when the assignation is enabled, compressing a second part of the point cloud data represented by a second tree of the second type wherein, features associated with a root node of the second tree are at least partially obtained from the given node.

Abstract: Systems, methods and apparatuses are described herein for controlling access to 3D media assets. An attribute of an entity represented in a 3D media asset may be identified, and a modified version of the 3D media asset may be generated by modifying such attribute. A request to access at least a portion of the 3D media asset may be received from a client device. A determination may be made, based on a policy associated with the 3D media asset, to enable access to the at least a portion of such asset. In response to such determination, an indication of how to process the modified version of the 3D media asset to reconstruct such asset comprising the attribute as identified prior to the modifying may be transmitted to the client device. Such client device may be caused to generate for display such 3D media asset based on the processing.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for generating marked digital images with content adaptive watermarks. In particular, in one or more embodiments, the disclosed systems intelligently evaluate a plurality of watermark configurations to select one or more content adaptive watermarks for one or more target digital images and generate one or more marked digital images by adding the selected content adaptive watermarks to the one or more target digital images.

Abstract: A robot control system includes a robot having a servo control unit, a lower-level control apparatus transmitting a control command to the servo control unit and receiving robot status information representing a status of the robot from the servo control unit with respect to each preset control period, and an upper-level control apparatus transmitting command information for creation of the control command to the lower-level control apparatus. The lower-level control apparatus transmits the robot status information to the upper-level control apparatus in synchronization with the control period. The upper-level control apparatus transmits the command information to the lower-level control apparatus within a predetermined transmission time shorter than the control period from a time when receiving the robot status information from the lower-level control apparatus.

Abstract: Methods, systems, and computer program products for stochastic compression of raster data are provided herein. A computer-implemented method includes obtaining at least one compression ratio and at least one error value for a given set of raster data; compressing at least a portion of the given set of raster data based at least in part on the at least one compression ratio and the at least one error value; transmitting the compressed raster data, to at least one given destination, based at least in part on a given transmission speed variable; and performing one or more automated actions based at least in part on the transmitted compressed raster data.

Abstract: Techniques are described for efficiently embedding frame masks in a video stream. In some solutions, a computer implemented method includes operations for encoding a frame of video data comprising an array of pixels to generate an encoded video frame, and transmitting the encoded video frame to a video decoder. The array of pixels include foreground pixels and background pixels. The foreground pixels have respective first chroma component values which are bounded within a first chroma component range. Encoding the frame of video data can include converting the first chroma component values of the foreground pixels to second chroma component values by applying a mapping function. The second chroma component values are bounded within a second chroma component range. The second chroma component range is shifted or compressed from the first chroma component range.

Abstract: A method, computer program, and computer system is provided for coding video data. Video data is received, and an edge present within a sample of the received video data is detected. A gradient value corresponding to a direction associated with the detected edge is calculated. The video data is decoded based on the calculated gradient.

Abstract: Methods, systems, and computer readable media for content item purging are provided. A contact item purger, such as may be incorporated within a local client application of a content management system running on a user device, may leverage knowledge as to which items have been uploaded to the content management system, and how long such content items have been stored on the user device, to propose items for deletion from the user device so as to reclaim storage space. A contact item purger may run on one or more user devices, and may activate upon various triggering events, based on various conditions and parameters, with or without user interaction, thus maintaining available memory capacity at all times.

Abstract: An arithmetic encoder for encoding a plurality of symbols having symbol values is configured to derive an interval size information for an arithmetic encoding of one or more symbol values to be encoded based on a plurality of state variable values representing statistics of a plurality of previously encoded symbol values with different adaptation time constants. The arithmetic encoder is configured to map a first state variable value, or a scaled and/or rounded version thereof, using a lookup-table and to map a second state variable value, or a scaled and/or rounded version thereof using the lookup-table, in order to obtain the interval size information describing an interval size for the arithmetic encoding of one or more symbols to be encoded. Further arithmetic encoders, arithmetic decoders, video encoders, video decoder, methods for encoding, methods for decoding and computer programs are also disclosed which are based on the same concept and on other concepts.

Abstract: An audio decoder for providing a decoded audio information includes a arithmetic decoder for providing a plurality of decoded spectral values on the basis of an arithmetically-encoded representation of the spectral values and a frequency-domain-to-time-domain converter for providing a time-domain audio representation using the decoded spectral values. The arithmetic decoder is configured to select a mapping rule describing a mapping of a code value onto a symbol code in dependence on a context state. The arithmetic decoder is configured to determine or modify the current context state in dependence on a plurality of previously-decoded spectral values. The arithmetic decoder is configured to detect a group of a plurality of previously-decoded spectral values, which fulfill, individually or taken together, a predetermined condition regarding their magnitudes, and to determine the current context state in dependence on a result of the detection. An audio encoder uses similar principles.

Abstract: A method, computer program, and computer system is provided for decoding point cloud data. Data corresponding to a point cloud is received. A number of contexts associated with the received data is reduced based on occupancy data corresponding to one or more parent nodes and one or more child nodes within the received data. The data corresponding to the point cloud is decoded based on the reduced number of contexts.

Abstract: A method, apparatus, article of manufacture, and a memory structure for signaling extension functions used in decoding a sequence comprising a plurality of pictures, each picture processed at least in part according to a picture parameter set is disclosed. In one embodiment, the method comprises reading a first extension flag signaling a first extension function in the processing of the sequence and determining if the first extension flag has a first value. Further, the method reads a second extension flag signaling a second extension function in the processing of the sequence and performs the second extension function according to the read second extension flag only if the first extension flag has a first value.

Abstract: A computing device performs initial processing of sensor data. The computing device includes one or more processors and instructions or logic that, when executed by the one or more processors, cause the computing device to perform obtaining sensor data, encoding the sensor data, writing the encoded sensor data to a dynamically allocated buffer, and logging a status of the written encoded sensor data at a static location of the dynamically allocated buffer. The status includes any one or more of memory addresses at which frames of the sensor data begin in the dynamically allocated buffer, valid bit fields corresponding to the frames, and sizes of each of data segments within the frames. The instructions further cause the computing device to perform, in response to receiving a polling request from a computing resource, transmitting the logged status to the computing resource.

Abstract: A processor obtains pieces of coded data, which are included in a bitstream and generated by coding tiles, and tile boundary independence information, which indicates whether each boundary between the tiles is a first or second boundary. Image data of a first tile is generated by decoding a first code string included in first coded data with reference to decoding information of a decoded tile when the tile boundary independence information indicates the first boundary, and by decoding the first code string without referring to the decoding information when the tile boundary independence information indicates the second boundary. A bit string is added after the first code string to make a bit length of first coded data a multiple of a predetermined N bits, with N being an integer greater than or equal to 2.

Abstract: The present invention enables a receiving side to easily recognize a high-quality format corresponding to encoded image data included in an extended video stream. Two video streams including a basic video stream including encoded image data of basic format image data, and an extended video stream including encoded image data of high-quality format image data of one type selected from a plurality of types are generated. A container of a predetermined format including the basic video stream and the extended video stream is transmitted. Information indicating a high-quality format corresponding to the encoded image data included in the extended video stream is inserted into the extended video stream and/or the container.

Abstract: A method, computer system, and computer program product are provided for managing content items, including tracking and/or updating content items. A content item is received from an author. A key is associated with the content item. Based on the key, a user is identified who is presenting the content item in a communication session. In response to determining that the author has updated the content item, the user is notified that an updated version of the content item is available for presentation in the communication session.

Abstract: The present disclosure relates to an image processing apparatus and an image processing method that are capable of improving coding efficiency at the time when a prediction using a correlation within a screen is performed. A predicted vector generation section sets, at a time of encoding of a current motion vector of a current block for a prediction using a correlation within a screen, a candidate block to not available and generates a predicted vector of the current motion vector by using a reference motion vector, the reference motion vector being referred to when the predicted vector of the current motion vector is generated. A difference vector generation section generates a difference vector between the current motion vector and the predicted vector generated by the predicted vector generation section. The present disclosure can be applied to, for example, an encoding apparatus.

Abstract: Techniques are described to automate deployment of an application as a cloud computing service. Such deployment may be accomplished on any underlying host infrastructure. As mentioned above, a typical application may engage in a number of build and deployment processes in order to run on a host computer with an infrastructure type. The techniques described herein may leverage an artificial intelligence computation library to identify which build and deployment processes are required by the application. Once identified, the user may select a host infrastructure type and create a container package with a runtime machine and components configure to run these build and deployment processes for that host infrastructure type. The container package may be used to generate an image from which the host computer can run the application as the cloud computing service. Other embodiments are described and claimed.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for channel-wise autoregressive entropy models. In one aspect, a method includes processing data using a first encoder neural network to generate a latent representation of the data. The latent representation of data is processed by a quantizer and a second encoder neural network to generate a quantized latent representation of data and a latent representation of an entropy model. The latent representation of data is further processed into a plurality of slices of quantized latent representations of data wherein the slices are arranged in an ordinal sequence. A hyperprior processing network generates a hyperprior parameters and a compressed representation of the hyperprior parameters. For each slice, a corresponding compressed representation is generated using a corresponding slice processing network wherein a combination of the compressed representations form a compressed representation of the data.