Patents by Inventor Kai Zhang

Kai Zhang has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • MATRIX DERIVATION IN INTRA CODING MODE
    Publication number: 20210385438
    Abstract: Devices, systems and methods for digital video coding, which includes matrix-based intra prediction methods for video coding, are described. In a representative aspect, a method for video processing includes performing a conversion between a current video block of a video and a bitstream representation of the current video block according to a rule, where the rule specifies a relationship between samples of the current video block and matrices or offset values applied in a matrix weighted intra prediction (MIP) mode during the conversion, and where the MIP mode includes determining a prediction block of the current video block by performing, on previously coded samples of the video, a boundary downsampling operation, followed by a matrix vector multiplication operation, and selectively followed by an upsampling operation.
    Type: Application
    Filed: August 11, 2021
    Publication date: December 9, 2021
    Inventors: Zhipin DENG, Kai ZHANG, Li ZHANG, Hongbin LIU, Jizheng XU
  • REGRESSION-BASED MOTION VECTOR FIELD BASED SUB-BLOCK MOTION VECTOR DERIVATION
    Publication number: 20210385483
    Abstract: Devices, systems and methods for digital video coding, which includes deriving RMVF-based motion vector information, are described. An exemplary method for video processing includes deriving at least one motion model for a current video block, based on motion information of at least one non-adjacent spatial neighboring block or at least one temporal neighboring block of the current video block; deriving motion information for the current video block or at least one sub-block of the current video block based on the at least one motion model; and performing a conversion for the current video block based on the derived motion information.
    Type: Application
    Filed: August 18, 2021
    Publication date: December 9, 2021
    Inventors: Hongbin LIU, Li ZHANG, Kai ZHANG, Jizheng XU, Yue WANG
  • Sub-block MV inheritance between color components
    Patent number: 11197007
    Abstract: Devices, systems and methods for sub-block based prediction are described. In a representative aspect, a method for video processing includes partitioning a first component of a current video block into a first set of sub-blocks and partitioning a second component of the current video block into a second set of sub-blocks. A sub-block of the second component corresponds to one or more sub-blocks of the first component. The method also includes deriving, based on a color format of the current video block, motion vectors for a sub-block of the second component based on motion vectors for one or more corresponding sub-blocks of the first color component.
    Type: Grant
    Filed: October 15, 2020
    Date of Patent: December 7, 2021
    Assignees: BEIJING BYTEDANCE NETWORK TECHNOLOGY CO., LTD., BYTEDANCE INC.
    Inventors: Kai Zhang, Li Zhang, Hongbin Liu, Yue Wang
  • Motion vector range based on motion vector precision
    Patent number: 11197016
    Abstract: A method for video processing is provided to comprise: determining a motion vector precision of one or more motion vectors related to a current video block of a video; selecting, based on the motion vector precision, a motion vector range; and performing, based on the motion vector range, a conversion between the current video block and a bitstream representation of the video.
    Type: Grant
    Filed: March 25, 2021
    Date of Patent: December 7, 2021
    Assignees: BEIJING BYTEDANCE NETWORK TECHNOLOGY CO., LTD., BYTEDANCE INC.
    Inventors: Li Zhang, Kai Zhang, Hongbin Liu, Jizheng Xu, Yue Wang
  • Unified constrains for the merge affine mode and the non-merge affine mode
    Patent number: 11197003
    Abstract: Devices, systems and methods for sub-block based prediction are described. In a representative aspect, a method for video processing comprises determining a block size constrain, making a determination, based on the block size constrain, about whether or not a merge affine mode and a non-merge affine mode are allowed for a video block in a video frame, and generating a bitstream representation of the video block based on the making the determination.
    Type: Grant
    Filed: November 16, 2020
    Date of Patent: December 7, 2021
    Assignees: BEIJING BYTEDANCE NETWORK TECHNOLOGY CO., LTD., BYTEDANCE INC.
    Inventors: Kai Zhang, Li Zhang, Hongbin Liu, Yue Wang
  • Mode dependent motion vector difference precision set
    Patent number: 11196999
    Abstract: A method of processing a video bitstream includes determining a motion precision set based on coding information of a current block. A conversion between a video block and a coded representation of the video block is performed based on the motion precision set. The conversion corresponds to a reconstruction of the current block. In some example aspects, a motion vector difference (MVD) precision of a current block from a motion precision set is determined based on a selected motion precision set and a MVD precision index. A conversion between a video block and a coded representation of the video block using an MVD is performed based on the MVD precision. The MVD represents a difference between a predicted motion vector and an actual motion vector used during motion compensation processing.
    Type: Grant
    Filed: July 28, 2020
    Date of Patent: December 7, 2021
    Assignees: BEIJING BYTEDANCE NETWORK TECHNOLOGY CO., LTD., BYTEDANCE INC.
    Inventors: Hongbin Liu, Li Zhang, Kai Zhang, Yue Wang
  • Gradient calculation in different motion vector refinements
    Patent number: 11197020
    Abstract: A video processing method is provided to include determining, for a current video block of a video, a final prediction block for the current video block by refining one or more initial predictions for the current video block using an optical flow calculation based on a gradient of initial prediction samples according to a precision rule; and performing a conversion between the current video block and a coded representation using the final prediction block, wherein the optical flow calculation includes a prediction refinement with optical flow (PROF) procedure or a bi-directional optical flow (BDOF) procedure, wherein the precision rule specifies to use a same precision for representing the gradient for both the PROF procedure and the BDOF procedure.
    Type: Grant
    Filed: August 3, 2021
    Date of Patent: December 7, 2021
    Assignees: BEIJING BYTEDANCE NETWORK TECHNOLOGY CO., LTD., BYTEDANCE INC.
    Inventors: Kai Zhang, Li Zhang, Hongbin Liu, Yue Wang
  • TRANSFORM PARAMETER DERIVATION BASED ON BLOCK PARTITION
    Publication number: 20210377531
    Abstract: Transform parameter derivation based on block partition is described. In an exemplary aspect, a method for video processing includes determining, for a conversion between a first block of video and a bitstream representation of the first block of video, whether a use of non-power-of-two partition trees (NPT-T) is enabled or disabled, wherein the NPT-T includes splitting the first block to multiple smaller sized child blocks of the first block, and at least one child block's width and/or height have a dimension that is a non-power-of-two integer; and performing the conversion based on the NPT-T in response to the determination that the NPT-T is enabled.
    Type: Application
    Filed: August 13, 2021
    Publication date: December 2, 2021
    Inventors: Li ZHANG, Kai ZHANG, Hongbin LIU, Yue WANG
  • RESTRICTIONS OF USAGE OF NON-POWER-TWO-PARTITION TREE IN VIDEO COMPRESSION
    Publication number: 20210377529
    Abstract: Restrictions of usage of non-power-two-partition tree in video compression is described. In an exemplary aspect, a method for video processing includes determining, for a conversion between a video and a bitstream representation of the video, whether non-power-of-two partition trees (NPT-T) partition is enabled or disabled, wherein the NPT-T partition include splitting a first block of the video into multiple smaller sized child blocks of the first block, and width (Wi) and/or height (Hi) of at least one child block is a non-power-of-two integer; determining restrictions associated with usage of the NPT-T partition in response to the determination that the NPT-T partition is allowed; and performing the conversion based on the determinations.
    Type: Application
    Filed: August 13, 2021
    Publication date: December 2, 2021
    Inventors: Li ZHANG, Kai ZHANG, Hongbin LIU, Yue WANG
  • INTERACTION BETWEEN INTRA BLOCK COPY MODE AND INTER PREDICTION TOOLS
    Publication number: 20210377561
    Abstract: The present disclosure relate to interaction between intra block copy mode and inter prediction tools A method for video processing is provided, including: determining that an Intra-Block-Copy (IBC) mode is applied to a current video block of a video, wherein in the IBC mode, at least one reference picture used by the current video block is a current picture where the current video block is located in; making a decision regarding a disabling of a specific coding mode for the current block; performing, based on the decision, a conversion between the current video block and the bitstream representation; wherein the specific coding mode uses a motion vector and a non-current picture to derive a prediction of a video block.
    Type: Application
    Filed: August 16, 2021
    Publication date: December 2, 2021
    Inventors: Kai ZHANG, Li ZHANG, Hongbin LIU, Yue WANG
  • INTERACTION BETWEEN LUT AND SHARED MERGE LIST
    Publication number: 20210377545
    Abstract: A method for processing video data is provided to include: determining whether a sharing of merge list information is enabled for a merge sharing node that corresponds to an ancestor node in a coding unit split tree to allow leaf coding units of the ancestor node in the coding unit split tree to use the merge list information; and performing a conversion between a current video block of a video and a bitstream of the video based on the determining.
    Type: Application
    Filed: July 13, 2021
    Publication date: December 2, 2021
    Inventors: Li ZHANG, Kai ZHANG, Hongbin LIU, Yue WANG
  • COORDINATION METHOD FOR SUB-BLOCK BASED INTER PREDICTION
    Publication number: 20210377521
    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.
    Type: Application
    Filed: August 16, 2021
    Publication date: December 2, 2021
    Inventors: Kai ZHANG, Li ZHANG, Hongbin LIU, Jizheng XU, Yue WANG
  • VIDEO BLOCK PARTITION BASED ON QUINARY-TREE
    Publication number: 20210377528
    Abstract: Devices, systems and methods for video processing are described. In a representative aspect, there is disclosed a method for video processing, including: determining, according to a splitting information, for a current video block of a video, a partition mode in which the current video block is split into M portions, wherein M>4; and performing a conversion for the current video block based on the partition mode.
    Type: Application
    Filed: August 10, 2021
    Publication date: December 2, 2021
    Inventors: Li ZHANG, Kai ZHANG, Hongbin LIU, Jizheng XU, Yue WANG
  • SELECTION OF CODED MOTION INFORMATION FOR LUT UPDATING
    Publication number: 20210377518
    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.
    Type: Application
    Filed: July 7, 2021
    Publication date: December 2, 2021
    Inventors: Li ZHANG, Kai ZHANG, Hongbin LIU, Yue WANG
  • NON-POWER-TWO-PARTITION TREE IN VIDEO COMPRESSION
    Publication number: 20210377530
    Abstract: Non-power-two-partition tree in video compression is described. In an exemplary aspect, a method for video processing includes determining, for a conversion between a first block of video and a bitstream representation of the first block of video, whether a use of non-power-of-two partition trees (NPT-T) is enabled or disabled, wherein the NPT-T includes splitting the first block to multiple smaller sized child blocks of the first block, and at least one child block's width and/or height have a dimension that is a non-power-of-two integer; and performing the conversion based on the NPT-T in response to the determination that the NPT-T is enabled.
    Type: Application
    Filed: August 13, 2021
    Publication date: December 2, 2021
    Inventors: Li ZHANG, Kai ZHANG, Hongbin LIU, Yue WANG
  • CLIPPING PARAMETER DERIVATION IN ADAPTIVE LOOP FILTER
    Publication number: 20210377524
    Abstract: Devices, systems and methods for temporal prediction of parameters in non-linear adaptive loop filtering are described. In an exemplary aspect, a method for visual media processing includes configuring, for a current video block, one or more parameters of a clipping operation that is part of a non-linear filtering operation; and performing, based on the one or more parameters, a conversion between the current video block and a bitstream representation of the current video block, wherein the one or more parameters is coded in accordance with a rule.
    Type: Application
    Filed: August 3, 2021
    Publication date: December 2, 2021
    Inventors: Li ZHANG, Kai ZHANG, Hongbin LIU, Yue WANG
  • SELECTIVE APPLICATION OF DECODER SIDE REFINING TOOLS
    Publication number: 20210377559
    Abstract: One example method of video processing includes implementing, by a processor, a decoder-side motion vector derivation (DMVD) scheme for motion vector refinement during a conversion between a current video block and a bitstream representation of the current video block by deriving parameters based on a deriving rule. The conversion may include compressing the current video block into the bitstream representation or uncompressing the bitstream representation into pixel values of the current video block.
    Type: Application
    Filed: August 13, 2021
    Publication date: December 2, 2021
    Inventors: Hongbin LIU, Li ZHANG, Kai ZHANG, Jizheng XU, Yue WANG
  • ROAD FEATURE POINT EXTRACTION METHOD AND SYSTEM
    Publication number: 20210365027
    Abstract: Disclosed are a road feature point extraction method and system. The road feature point extraction method comprises: acquiring map information for the current position of a vehicle, wherein the map information comprises the attributes of the current road and the attributes of the next road section; comparing the attributes of the current road with the attributes of the next road section to determine the road type of the next road section; and extracting, in conjunction with the road type of the next road section and the map information corresponding to the next road section, a road feature point representing a road scene on the next road section, and outputting the name of the extracted road feature point and information of the relative distance between the road feature point and the vehicle.
    Type: Application
    Filed: September 27, 2019
    Publication date: November 25, 2021
    Inventors: Lin HE, Longbao ZHEN, Kai ZHANG, Junpeng ZHAO, Wei LI, Tianpei WANG, Hongliang LIU, Ronglin ZENG, Weifeng DENG
  • GRADIENT CALCULATION IN DIFFERENT MOTION VECTOR REFINEMENTS
    Publication number: 20210368203
    Abstract: A video processing method is provided to include determining, for a current video block of a video, a final prediction block for the current video block by refining one or more initial predictions for the current video block using an optical flow calculation based on a gradient of initial prediction samples according to a precision rule; and performing a conversion between the current video block and a coded representation using the final prediction block, wherein the optical flow calculation includes a prediction refinement with optical flow (PROF) procedure or a bi-directional optical flow (BDOF) procedure, wherein the precision rule specifies to use a same precision for representing the gradient for both the PROF procedure and the BDOF procedure.
    Type: Application
    Filed: August 3, 2021
    Publication date: November 25, 2021
    Inventors: Kai ZHANG, Li ZHANG, Hongbin LIU, Yue WANG
  • DIRECTION-BASED PREDICTION FOR INTRA BLOCK COPY IN VIDEO CODING
    Publication number: 20210368178
    Abstract: A method of visual media processing includes determining, for a conversion between a current video block of visual media data and a bitstream representation of the current video block, a buffer that stores reference samples for prediction in an intra block copy mode; for a sample spatially located at location of the current video block relative to an upper-left position of a coding tree unit including the current video block and having a block vector, computing a corresponding reference in the buffer at a reference location, wherein the reference location is determined using the block vector and the location; and upon determining that the reference location lies outside the buffer, re-computing the reference location based at least in part on a location of the current video block relative to the coding tree unit including the current video block.
    Type: Application
    Filed: June 29, 2021
    Publication date: November 25, 2021
    Inventors: Jizheng XU, Li ZHANG, Kai ZHANG, Hongbin LIU, Yue WANG