Abstract: This invention describes a method for communicating crude motion information using tracking metadata and recovering more accurate motion information from the received tracking metadata and partial video frame data; in particular, we use metadata to convey crude boundaries of objects in the scene and signal motion information for these objects. The proposed method leaves the task of identifying the exact boundaries of an object to the decoder/client. The proposed method is particularly appealing when metadata itself carries semantics that the client is interested in, such as tracking information in surveillance applications, because, in this case, metadata does not constitute an overhead. The proposed method involves motion descriptions that can be used to predict the appearance of an object in any one frame from its appearance in any other frame that contains the object.

Abstract: Devices, systems and methods for processing video are described. In a representative aspect, a video processing method is provided to comprise: maintaining tables, wherein each table includes a set of motion candidates and each motion candidate is associated with corresponding motion information; performing a conversion between a first video block and a bitstream representation of a video including the first video block based on the tables; and updating, after performing of the conversion, zero or more tables, based on an update rule.

Abstract: The invention relates to a method for encoding/decoding an image. The image decoding method according to the invention comprises the steps of: obtaining diagonal partition information on a current block; determining a diagonal partition structure of the current block using the diagonal partition information; and diagonally partitioning the current block into a first and a second area based on the determined diagonal partition structure, the current block being a leaf node of a square or rectangular partition.

Abstract: Disclosed are an image coding and decoding methods, an image processing device, and a computer storage medium. The image coding method comprises: obtaining a dynamic range of a copy parameter of a current coding sampling value segment according to a size of a reference area; and coding the copy parameter according to the dynamic range to generate a video bitstream containing information about the size of the reference area and information about the copy parameter. The decoding method comprises: parsing a video bitstream to obtain a dynamic range of a copy parameter of a decoding sampling value segment; and decoding the copy parameter according to the dynamic range.

Abstract: A motion vector determination employs template matching. At first, an initial motion vector is obtained. If the initial motion vector points to a position that is fractional, and thus, requires interpolation from the integer sample positions, the initial motion vector is rounded to a closest integer sample position. The rounded position is then used to define the search space for the template matching to define a refinement of the initial motion vector.

Abstract: A method of video decoding includes maintaining a number of tables, wherein each table includes a set of motion candidates and each motion candidate is associated with corresponding motion information derived from previously coded video blocks, performing a conversion between a current video block and a bitstream representation of the current video block in a video region, and updating the one or more tables based on an updating rule.

Abstract: A method includes receiving a reference image and a non-reference image; dividing the reference image into a plurality of tiles; determining, using an electronic device, a motion vector map using coarse-to-fine based motion vector estimation; and generating an output frame using the motion vector map with the reference image and the non-reference image.

Abstract: The present disclosure relates to motion vector determination using template matching. In particular, an initial motion vector and a template are obtained for a prediction block. Then a refinement of the initial motion vector is determined by template matching with said template in a search space. The search space is located on a position given by the initial motion vector. The template matching accesses integer sample positions within a predetermined window, the predetermined window including said search space and further integer sample positions accessible for the template matching, replaces at least one integer sample position located out of the window by at least one replacement value, each replacement value is based on at least one sample within said window, and uses the accessed integer sample position values as well as the at least one replacement value to perform the template matching.

Abstract: A video encoder or video decoder can be configured to obtain multi-layer video data that includes at least a first layer and a second layer; determine a layer identification value for the first layer and a layer identification value for the second layer, wherein the layer identification value for the second layer is higher than the layer identification value for the first layer; store decoded pictures of the first layer and decoded pictures of the second layer in a decoded picture buffer; and in response to determining that a picture of a layer is an instantaneous decoding refresh (IDR) picture, remove from the decoded picture buffer, all decoded pictures of a layer the IDR picture belongs to, while leaving in the decoded picture buffer all decoded pictures with layer identifications lower than a layer identification of the layer the IDR picture belongs to.

Abstract: An encoding device includes: an index setting unit that sets an index that represents information of a reference image and a weighting factor; an index reconfiguring unit that predicts a reference value of the weighting factor, wherein the reference value indicates a factor to be set if a difference of pixel value between a reference image and a target image to be encoded is less than or equal to a specific value; and an entropy encoding unit that encodes a difference value between the weighting factor and the reference value.

Abstract: Techniques are provided for synthesis of transformed image views, based on a reference image, using depth information. The transformed image views may simulate a change in position or focal length of a camera that produced the reference image. An example system includes an image transformation circuit configured to transform the reference image corresponding to a first viewpoint, to a transformed image corresponding to a second viewpoint. The system also includes an inverse warping circuit configured to calculate a mapping from the pixels of the transformed image to corresponding pixels of the reference image. The system further includes a hole detection circuit configured to detect holes in the transformed image based on depth discontinuities between the reference and transformed images; and a hole filling circuit configured to in-fill the detected holes using a sampling of selected neighboring pixels from the reference image, to synthesize a view based on the transformed image.

Abstract: Techniques for implementing video processing techniques are described. In one example implementation, a method of video processing includes determining, for a conversion between a current block of a video and a bitstream representation of the video, whether a combined inter and intra prediction (CIIP) coding technique is to be applied to the current block based on a characteristic of the current block. The CIIP coding technique uses an intermedia inter prediction value and an intermedia intra prediction value to derive a final prediction value of the current block. The method also includes performing the conversion based on the determining.

Abstract: An encoder includes memory and circuitry. The circuitry: derives a first motion vector in a unit of a prediction block using a first inter frame prediction mode that uses a degree of matching between two reconstructed images of two regions in two difference pictures, the prediction block being obtained by splitting an image included in a video; and performs, in the unit of the prediction block, a first motion compensation process that generates a prediction image by referring to a spatial gradient of luminance in an image generated by performing motion compensation using the first motion vector derived.

Abstract: A method of partitioning a video coding block for JVET, comprising representing a JVET coding tree unit as a root node in a quadtree plus binary tree (QTBT) structure that can have a quadtree branching from the root node and binary trees branching from each of the quadtree's leaf nodes using asymmetric binary partitioning to split a coding unit represented by a quadtree leaf node into two child coding units of unequal size, representing the two child coding units as leaf nodes in a binary tree branching from the quadtree leaf node and coding the child coding units represented by leaf nodes of the binary tree with JVET, wherein further partitioning of child coding units split from quadtree leaf nodes via asymmetric binary partitioning is disallowed.

Abstract: Aspects of the disclosure provide methods and apparatuses for video encoding/decoding. In some examples, an apparatus for video decoding includes receiving circuitry and processing circuitry. In some embodiments, the processing circuitry decodes prediction information of a current block in a current coding from a coded video bitstream. The prediction information is indicative of an intra block copy mode. Then, the processing circuitry determines a block vector that points to a reference block in a same picture as the current block. The reference block is restricted within a coding region with reconstructed samples buffered in a reference sample memory. The coding region is one of multiple predefined regions of a coding tree unit (CTU). Then, the processing circuitry reconstructs at least a sample of the current block based on the reconstructed samples of the reference block that are retrieved from the reference sample memory.

Abstract: A method for inter-coding video is provided in which transmission bandwidth requirements associated with second motion vectors for bi-directional temporal prediction is reduced. In the method motion vector information for only one of the two motion vectors for bi-directional temporal prediction can be transmitted together with information on how to derive or construct the second motion vector. Thus, rather than sending express information regarding two motion vectors, express information related to only one motion vector along with information related to reconstruction/derivation of the second motion vector is transmitted, thus reducing bandwidth requirements and increasing coding efficiency.

Abstract: A system and method of reducing blocking artifacts and providing enhanced coding efficiency based, as least in part, upon evaluation of relative smoothness of signals at a coding boundary. In some embodiments, a boundary threshold difference can be established beyond which it is determined that the difference is representative of a natural or intended boundary and filtering can be applied to those boundaries having differences below the boundary threshold difference. In some further embodiments, the ramps of the signal across the boundary can be evaluated to determine whether weak or strong filtering might be appropriate. In some further embodiments, weak filtering can be performed that reduces blocking artifacts, improves coding efficiency, but does not distort ramp signals across the boundary.

Abstract: A sequence of digital images is encoded into a bitstream, at least one portion of an image being encoded by motion compensation with respect to a reference image portion. A target number of motion information predictors is obtained. Using the target number a set of motion information predictors is generated having controlled diversity. A motion information predictor for the image portion to encode is selected from the generated set of motion information predictors. The target number is signaled in the bitstream, together with information relating to the selected motion information predictor.

Abstract: A method of video decoding includes maintaining a number of tables, wherein each table includes a set of motion candidates, wherein each motion candidate is associated with corresponding motion information derived from previously video blocks, performing a conversion between a video block and a coded representation of the video block, and determining, based on a conversion condition of the video block, whether to update at least one of the tables by adding motion information corresponding to the video block.

Abstract: The present disclosure relates to the search of more than one K-integer best-matching blocks per block within an image, corresponding to best patches for subsequent filtering. In particular, the positions of K best-matching blocks for a template block are found within an image search area, by performing calculations of the similarity between the template block and a test block at all offset positions within a search area. The positions of K or more best-matching blocks for a non-template block are found within an image search area, by performing calculations of the similarity between the non-template block and a test block at all offset positions found as offsets of best-matching blocks for all template blocks.

Abstract: A method of video processing includes maintaining tables, wherein each table includes a set of motion candidates and each motion candidate is associated with corresponding motion information; and performing a conversion between a first video block and a bitstream representation of a video including the first video block based on at least some of tables, the performing including checking motion candidates in a table in an order defined in a rule until a checking termination criteria is met.

Abstract: A method for decoding a video according to the present invention may comprise: determining a first prediction mode for a first sub-block in a current block and a second intra prediction mode for a second sub-block, performing a first prediction for the first sub-block based on the first intra prediction mode, performing a second prediction for the second sub-block based on the second intra prediction mode, and obtaining a prediction sample of the current block according to a result of the first prediction and the second prediction.

Abstract: A method of decoding an encoded video bitstream using at least one processor, including obtaining a first flag indicating whether a constant picture size is used in a coded video sequence including a current picture; based on the first flag indicating that the constant picture size is used, decoding the current picture without performing reference picture resampling; based on the first flag indicating that the constant picture size is not used, obtaining a second flag indicating whether a conformance window size is signaled; based on the second flag indicating that the conformance window size is signaled: obtaining the conformance window size, determining a resampling ratio between the current picture and a reference picture based on the conformance window size, and performing the reference picture resampling on the current picture using the resampling ratio.

Abstract: Devices, systems and methods for digital video coding, which include sub-block based inter prediction methods, are described. An exemplary method for video processing includes determining, for a conversion between a current block of video and a bitstream representation of the video, a maximum number of candidates in a sub-block based merge candidate list and/or whether to add sub-block based temporal motion vector prediction (SbTMVP) candidates to the sub-block based merge candidate list based on whether temporal motion vector prediction (TMVP) is enabled for use during the conversion or whether a current picture referencing (CPR) coding mode is used for the conversion, and performing, based on the determining, the conversion.

Abstract: An encoding device includes an index setting unit that sets an index that represents information of a reference image and a weighting factor. An index reconfiguring unit predicts a reference value of the weighting factor. The reference value indicates a factor to be set if a difference of pixel value between a reference image and a target image to be encoded is less than or equal to a specific value. An entropy encoding unit encodes a difference value between the weighting factor and the reference value.

Abstract: A method of coding prediction information of a current block includes determining that the current block is to be coded according to an IntraBC mode (Intra Block Copy mode) and determining one or more block vector (BV) prediction candidates from one or more motion vectors (MVs) or BVs associated with neighboring blocks of the current block according to an advanced motion vector prediction (AMVP) mode or a Merge/Skip mode. The method further includes encoding or decoding a current BV based on a BV predictor selected from the one or more BV prediction candidates, where the current BV is for coding the current block according to the IntraBC mode.

Abstract: A method for video processing is provided to include constructing, during a conversion between a current video block and a coded representation of the current video block, an intra block copy (IBC) motion candidate list, wherein insertion of motion candidates to the IBC motion candidate list is subject to a size restriction; and performing the conversion using the IBC motion candidate list.

Abstract: Systems and techniques for improving costs and efficiency in affine motion prediction of a current block includes storing motion information of previously coded block in a line buffer or one or more local buffers of a coding unit such as coding tree unit (CTU). The line buffer can store motion information of a line of coding blocks located at a boundary of the CTU, while the local buffers can be utilized for storing motion information of coding blocks located anywhere in the CTU. The utilization of the line buffer can be improved by limiting the number of motion vectors of a neighboring block stored in the line buffer and obtaining remaining motion vectors from an additional block when available. The utilization of local buffers can be improved by restricting a size of neighboring blocks whose motion information is stored in the local buffers.

Abstract: A method of video processing is provided to include maintaining one or more tables, wherein each table includes one or more motion candidates and each motion candidate is associated with corresponding motion information; performing a conversion between a current block and a bitstream representation of a video including the current block by using motion information in a table; and updating, after performing of the conversion, one or more tables based on M sets of additional motion information associated with the current block, M being an integer.

Abstract: A decoding method of multi-view video information is provided. The decoding method of multi-view video information includes: receiving a bit stream about video data of a plurality of videos photographed by a plurality of cameras; generating first information by extracting depth information of a block in a video frame of a plurality of videos from video data included in the bit stream; generating second information by determining whether the same motion estimation is generated between video frames of the plurality of videos; generating third information by extracting pixel information of a block in the video frame of the plurality of videos; and performing block matching and disparity compensation between video frames of the plurality of videos by inputting the first information to third information to a disparity compensation model of multi-view video information which is generated by being trained in advance.

Abstract: High-quality video is provided using a small amount of coded bits. The moving picture decoding method performs inter-frame prediction processing. With the aforementioned inter-frame prediction processing, blocks with similar motion vectors from among the motion vectors in multiple blocks that have already been decoded are combined and a combined area is computed. A predicted vector for a target block to be decoded is computed using the motion vector of the aforementioned combined area, and a motion vector for the aforementioned target block is computed based on the aforementioned predicted vector and a difference vector which is included in a coded stream that is input. A predicted image is generated using the aforementioned motion vector, and a difference image which is included in the aforementioned coded stream and the aforementioned predicted image are added to generate a decoded image.

Abstract: Devices, systems and methods for digital video coding, which include decoder side motion vector derivation (DMVD) tools, are described. An exemplary method for video processing includes making a decision, based on a determination that a current block of a video is coded using a multi-hypothesis prediction mode, regarding a selective enablement of a DMVD tool for the current block, wherein the DMVD tool derives a refinement of motion information signaled in a bitstream representation of the video; and performing, based on the decision, a conversion between the current block and the bitstream representation.

Abstract: An encoder that encodes a picture to generate a coded stream includes: circuitry and a memory coupled to the circuitry. The circuitry performs, using the memory: generating a prediction image of a current block included in a current picture by referring to a first region included in a reference picture different from the current picture; operating a bi-directional optical flow process to correct the prediction image by referring to a second region included in the first region, and not operating the bi-directional optical flow process in response to the second region not being included in the first region; and encoding the current block based on the prediction image.

Abstract: Devices, systems and methods for encoding and decoding digital video historical information such as tables containing coding candidates are described. In a representative aspect, a method for video processing includes maintaining an Advanced Motion Vector Prediction (AMVP) candidate list for a conversion between a video block of a video and a bitstream representation of the video with a motion vector precision of M-Pel or sub-pel selected from multiple available motion vector precisions, M being a positive integer. The method also includes comparing an AMVP candidate associated with the video block with existing AMVP candidates in the AMVP candidate list, adding the AMVP candidate to the AMVP candidate list upon determining that the AMVP candidate is different than the existing candidates, and performing the conversion based on the AMVP candidate list.

Abstract: A device for processing video data includes a memory configured to store video data and one or more processors implemented in circuitry. The one or more processors are configured to generate a coding unit for chroma components of a block of video data. The one or more processors are configured to split the coding unit for chroma components into a first triangle-shaped partition and a second triangle-shaped partition. The one or more processors are configured to apply pixel blending using a set of weights for a YUV 4:2:0 format to generate a predicted block for the chroma components of the block of video data when the one or more processors generate the coding unit for chroma components in the YUV 4:2:0 format and when the one or more processors generate the coding unit for chroma components in a YUV 4:4:4 format.

Abstract: Methods and apparatuses are provided for video coding. The method includes: partitioning video pictures into a plurality of coding units (CUs), at least one of which is further partitioned into two prediction units (PUs) including at least one triangular shaped PU; constructing a first merge list including a plurality of candidates, each including one or more motion vectors, based on a merge list construction process for regular merge prediction; and obtaining an index listing including a plurality of reference indices, where each reference index comprises a reference to a motion vector of a candidate in the first merge list.

Abstract: A method for setting the motion vector list and the apparatus using the same may include determining the presence of a first motion vector or a second motion vector by a sequential determination process in a first spatial candidate prediction group; and setting the first motion vector or the second motion vector produced through the sequential determination process as the candidate prediction motion vector. Thus, the encoding/decoding time and the complexity can be reduced by restricting the scaling number in a process for scanning the candidate prediction motion vector.

Abstract: Techniques are disclosed relating to operand routing among SIMD pipelines. In some embodiments, an apparatus includes a set of multiple hardware pipelines configured to execute a single-instruction multiple-data (SIMD) instruction for multiple threads in parallel, wherein the instruction specifies first and second architectural registers. In some embodiments, the pipelines include execution circuitry configured to perform operations using one or more pipeline stages of the pipeline. In some embodiments, the pipelines include routing circuitry configured to select, based on the instruction, a first input operand for the execution circuitry from among: a value from the first architectural register from thread-specific storage for another pipeline and a value from the second architectural register from thread-specific storage for a thread assigned to another pipeline.

Abstract: The present disclosure describes a system and method for background identification for videos with large foreground objects. The method includes dividing a video clip into a multiple short video clips, each of which overlaps with neighboring short video clips of the multiple short video clips, detecting, in each short video clip, multiple background motion candidates, linking the background motion candidates between the neighboring short video clips to form multiple linked groups of background motion candidates, and identifying a linked group of background motion candidates that is most dominant among the multiple linked groups of background motion candidates.

Abstract: Aspects of the disclosure provide method and apparatus for video coding. In some examples, an apparatus includes processing circuitry for video decoding. The processing circuitry decodes prediction information of a current block from a coded video bitstream. The prediction information is indicative of an intra block copy mode. The processing circuitry selects, from a set of multiple candidate resolutions, a resolution of a block vector difference for the current block, and determines a block vector of the current block according to the selected resolution of the block vector difference and a block vector predictor of the current block. Then, the processing circuitry reconstructs at least one sample of the current block according to the block vector.

Abstract: The architecture shown can perform global search, local search and local sub pixel search in a parallel or in a pipelined mode. All operations are in a streaming mode without the requirement of external intermediate data storage.

Abstract: A method of and an apparatus for controlling intra prediction for decoding of a video sequence are provided. The method includes setting a table including a plurality of intra prediction modes respectively corresponding to a plurality of intra prediction angles, the plurality of intra prediction angles including diagonal intra prediction angles respectively corresponding to diagonal directions of all shapes of coding units, and selecting, for decoding the video sequence, one of the plurality of intra prediction modes, from the set table.

Abstract: A method of video processing is provided to include determining, for a conversion between a current video block of a video and a coded representation of the video, a mode of operation of an UMVE mode; and performing the conversion based on the determining, wherein the current video block is coded with a merge mode and motion vector differences in an UMVE mode that comprises a motion vector expression that includes a starting point of motion information, a motion magnitude and a motion direction for the current video block, and wherein one or more fields in the coded representation correspond to the mode of operation, and wherein the one or more fields include: a list size field that indicates a size of a base candidate list used by the UMVE mode, or a table field that signals a distance table or a direction table for the UMVE mode.

Abstract: A method and apparatus of Inter prediction for video coding including IBC (Intra Block Copy) are disclosed. In one method, an IBC Merge candidate list is generated from Merge candidates associated with one or more candidate types, wherein one or more target candidates associated with a target candidate type are excluded from the IBC Merge candidate list when one or more constraints are satisfied. Current motion information associated with the current block is encoded or decoded using the IBC Merge candidate list. According to another method, a pruning process for the IBC Merge candidate list is skipped or simplified according to block width, block height, or block area of the current block.

Abstract: An apparatus and a method for motion estimation of isolated objects are provided. The apparatus includes a dividing circuit, a flag circuit, an analyzing circuit, a calculating circuit and a determination circuit. The dividing circuit divides each of a first image and a second image into a plurality of blocks. The flag circuit adds a flag to each block within the blocks having image information different from background information. The analyzing circuit analyzes continuity of the flag to find a target isolated object having a size less than or equal to a predetermined size. The calculating circuit obtains an initial motion vector of the target isolated object according to locations of the target isolated object in the first image and the second image. The determination circuit determines whether the initial motion vector is correct according to a sum of absolute differences corresponding to the initial motion vector.

Abstract: A motion vector determination employs template matching. At first, an initial motion vector is obtained. If the initial motion vector points to a position that is fractional, and thus, requires interpolation from the integer sample positions, the initial motion vector is rounded to a closest integer sample position. The rounded position is then used to define the search space for the template matching to define a refinement of the initial motion vector.

Abstract: An image signal decoding method according to the present invention comprises the steps of: decoding information indicating whether a current block is encoded using a multi-mode intra prediction; when it is determined that the current block is encoded in the multi-mode intra prediction, dividing the current block into a plurality of partial blocks; and obtaining an intra prediction mode of each of the plurality of partial blocks.

Abstract: The present disclosure relates to motion vector determination using template matching. In particular, an initial motion vector and a template are obtained for a prediction block. Then a refinement of the initial motion vector is determined by template matching with said template in a search space. The search space is located on a position given by the initial motion vector. The template matching accesses integer sample positions within a predetermined window, the predetermined window including said search space and further integer sample positions accessible for the template matching, replaces at least one integer sample position located out of the window by at least one replacement value, each replacement value is based on at least one sample within said window, and uses the accessed integer sample position values as well as the at least one replacement value to perform the template matching.

Abstract: The present disclosure relates to a method and apparatus for determining a target rotation direction. Said method for determining the target rotation direction comprises: inputting successive video frames which comprise a rotation target; establishing a background model for the first image frame in said video frames; performing foreground detection on the video frames other than the first frame by means of said background model so as to determine the rotation axis of said rotation target; obtaining the distribution of optical flow points within a preset region of said rotation axis; determining the direction of rotation of said rotation target according to the distribution of optical flow points within said preset region. By means of the present disclosure, it is possible to simply and efficiently determine the clock direction of a rotation target in a video.

Abstract: A method and apparatus are provided for encoding and decoding image information. The encoding comprises receiving a block of pixels; creating a set of motion vector prediction candidates for the block of pixels; and examining the set to determine if a motion vector prediction candidate is a temporal motion vector prediction, or a spatial motion vector prediction. If the motion vector prediction candidate is a temporal motion vector prediction, the motion vector prediction candidate is kept in the set. If the motion vector prediction candidate is a spatial motion vector prediction, it is examined whether the set comprises a motion vector prediction candidate corresponding with the spatial motion vector prediction; and if so, the motion vector prediction candidate is removed from the set. Once the set is created, one of the candidates from the set is selected to represent a motion vector prediction for the block of pixels.