Patents by Inventor Xinwei Li

Xinwei Li 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).

  • Publication number: 20250227241
    Abstract: Methods and systems implement application of template-matching-based motion refinement on subblocks of coding blocks. A VVC-standard encoder and a VVC-standard decoder can configure one or more processors of a computing system to obtain sub-templates of each subblock of a coding block from respective motion vectors of neighboring subblocks.
    Type: Application
    Filed: January 2, 2025
    Publication date: July 10, 2025
    Inventors: Jie Chen, Ru-ling Liao, Yan Ye, Xinwei Li
  • Publication number: 20250227292
    Abstract: Methods and systems implement fusion of intra TMP mode with other intra prediction modes that utilize adjacent samples, to improve prediction accuracy. A VVC-standard encoder and a VVC-standard decoder can configure one or more processors of a computing system to apply non-CCP modes on the template of the collocated luma block, and reorder non-CCP modes based on template matching cost; additionally reorder angular modes, such as a subset of efficient angular modes based on the template of the collocated luma block; prune non-CCP modes from the ordered list based on similarity; move a non-CCP mode of the reordered ordered list based on template matching cost difference relative to a predecessor; copy and reorder the ordered list of non-CCP modes once for each respective distinctly signaled chroma fusion mode; fuse a chroma DBV mode with a CCP mode; and select a least-cost reordered block vector for chroma DBV mode.
    Type: Application
    Filed: January 2, 2025
    Publication date: July 10, 2025
    Inventors: Xinwei Li, Ru-ling Liao, Jie Chen, Yan Ye
  • Patent number: 12355982
    Abstract: A VVC-standard encoder and a VVC-standard decoder are provided, implementing a decoder-side chroma intra prediction mode gradient-based derivation method, which improves coding efficiency of chroma intra prediction, thereby saving on signaling cost. A VVC-standard decoder configures one or more processors of a computing system to derive one of multiple possible chroma intra prediction modes by computing gradients of adjacent luma samples and chroma samples of a current chroma block. With minimal increase in signaling cost, the VVC-standard coding and decoding processes are enhanced to base intra prediction modes for chroma blocks on collocated luma block-adjacent reconstructed luma samples adjacent reconstructed chroma samples, enabling prior computational work done on coding and decoding adjacent blocks to be referenced. In this fashion, coding gains can be achieved in matching texture characteristics of the current chroma block based on a texture gradient including adjacent blocks.
    Type: Grant
    Filed: December 29, 2022
    Date of Patent: July 8, 2025
    Assignee: Alibaba (China) Co., Ltd
    Inventors: Xinwei Li, Ru-Ling Liao, Jie Chen, Yan Ye
  • Patent number: 12347461
    Abstract: The present disclosure provides a video generation method, an apparatus, a device, and a storage medium, and the method includes: displaying shooting indication information of storyboard shots corresponding to a target video editing template in response to a trigger operation for the target video editing template, the shooting indication information being configured to indicate shooting modes and/or shooting content features corresponding to the storyboard shots; acquiring user material for the storyboard shots corresponding to the target video editing template, respectively; and generating a target result video corresponding to the target video editing template according to the user material respectively corresponding to the storyboard shots.
    Type: Grant
    Filed: December 4, 2023
    Date of Patent: July 1, 2025
    Assignee: Beijing Zitiao Network Technology Co., Ltd.
    Inventors: Xinwei Li, Jianfan Zhao
  • Patent number: 12348724
    Abstract: The present disclosure provides methods and systems for fusing chroma intra prediction modes. An exemplary method includes: generating a plurality of predicted chroma samples associated with a pixel, by using a plurality of chroma intra prediction modes respectively; and determining a first predicted chroma sample, based on a weighted sum of the plurality of predicted chroma samples.
    Type: Grant
    Filed: December 23, 2022
    Date of Patent: July 1, 2025
    Assignee: Alibaba Innovation Private Limited
    Inventors: Xinwei Li, Ru-Ling Liao, Jie Chen, Yan Ye
  • Publication number: 20250211730
    Abstract: A transform method and a decoder. The encoder determines a prediction mode parameter of a current block; determines a cross-component linear model (CCLM) parameter when the prediction mode parameter indicates that CCLM prediction is used for the current block to determine an intra prediction value; determines the intra prediction value of the current block according to the CCLM parameter, and calculates a residual value between the current block and the intra prediction value; determines a low-frequency non-separable transform (LFNST) transform kernel used for the current block according to the CCLM parameter, sets an LFNST index, and signals the LFNST index into a bitstream of a video, when LFNST is used for the current block; and transforms the residual value with the LFNST transform kernel.
    Type: Application
    Filed: March 7, 2025
    Publication date: June 26, 2025
    Inventors: Junyan HUO, Xinwei LI, Wenhan QIAO, Yanzhuo MA, Shuai WAN, Fuzheng YANG
  • Publication number: 20250203089
    Abstract: The present disclosure provides a video data processing method for cross-component sample adaptive offset (CCSAO). The method includes determining an index based on a vertical coordinate of a chroma sample within a picture; determining a luma sample based on the index; classifying the chroma sample based on a reconstructed value associated with the luma sample; determining an offset based on the classification; and adding the offset to a reconstructed value associated with the chroma sample.
    Type: Application
    Filed: February 27, 2025
    Publication date: June 19, 2025
    Inventors: Xinwei LI, Jie CHEN, Ru-Ling LIAO, Yan YE
  • Publication number: 20250193392
    Abstract: The present disclosure provides a video data processing method. The method includes: receiving a bitstream; decoding a first index from the bitstream; determining a maximum number of an adaptive loop filter (ALF) for a component of a picture based on the first index; and processing pixels in the picture with the ALF.
    Type: Application
    Filed: February 24, 2025
    Publication date: June 12, 2025
    Inventors: Jie CHEN, Ru-Ling LIAO, Xinwei LI, Yan YE
  • Publication number: 20250193400
    Abstract: A method for video data processing includes: determining a category index of a target chroma sample, wherein the category index is determined based on a first reconstructed value associated with a co-located luma sample and a second reconstructed value associated with the target chroma sample; determining an offset based on the category index; and adding the offset to a third reconstructed value associated with the target chroma sample.
    Type: Application
    Filed: February 14, 2025
    Publication date: June 12, 2025
    Inventors: Xinwei LI, Ru-Ling LIAO, Jie CHEN, Yan YE
  • Publication number: 20250159200
    Abstract: Chroma intra prediction methods and devices are provided. The method includes: determining a direct mode (DM) corresponding to a current chroma block; determining an intra prediction mode of the current chroma block from a set of chroma prediction modes, wherein the set of chroma prediction modes comprises at least one of the DM, linear mode (LM), LM_T or LM_L, wherein when a luma prediction mode corresponding to the DM is a direct current (DC) mode or a planar mode, the set of chroma prediction modes comprises a diagonal angular mode with prediction direction index number 66, wherein the linear mode is acquired from a bitstream before determining the set of chroma prediction modes; and determining a reconstructed value of the current chroma block according to the intra prediction mode.
    Type: Application
    Filed: January 16, 2025
    Publication date: May 15, 2025
    Inventors: Junyan HUO, Yanzhuo MA, Shuai WAN, Fuzheng YANG, Xinwei LI
  • Publication number: 20250133203
    Abstract: A prediction method, apparatus, and a computer storage medium for encoding, the method includes: acquiring reference samples adjacent to at least one side of a coding block; determining a reference point from the at least one side and determining reference sample positions corresponding to the at least one side according to a preset number of samples; deriving reference samples corresponding to the reference sample positions from the reference samples; and performing prediction on the coding block based on the derived reference samples.
    Type: Application
    Filed: December 23, 2024
    Publication date: April 24, 2025
    Inventors: Junyan HUO, Yanzhuo MA, Shuai WAN, Fuzheng YANG, Xinwei LI
  • Patent number: 12278948
    Abstract: A transform method and a decoder. The encoder determines a prediction mode parameter of a current block; determines a cross-component linear model (CCLM) parameter when the prediction mode parameter indicates that CCLM prediction is used for the current block to determine an intra prediction value; determines the intra prediction value of the current block according to the CCLM parameter, and calculates a residual value between the current block and the intra prediction value; determines a low-frequency non-separable transform (LFNST) transform kernel used for the current block according to the CCLM parameter, sets an LFNST index, and signals the LFNST index into a bitstream of a video, when LFNST is used for the current block; and transforms the residual value with the LFNST transform kernel.
    Type: Grant
    Filed: June 22, 2022
    Date of Patent: April 15, 2025
    Assignee: GUANGDONG OPPO MOBILE TELECOMMUNICATIONS CORP., LTD.
    Inventors: Junyan Huo, Xinwei Li, Wenhan Qiao, Yanzhuo Ma, Shuai Wan, Fuzheng Yang
  • Publication number: 20250119627
    Abstract: The disclosure relates to a method, apparatus, device and medium for producing video. The method includes: creating a video producing task for a user group comprising at least a first user and a second user, the video producing task being used for collecting a storyboard video material and/or editing the storyboard video material into a target video; in response to a first editing operation of the first user for the video producing task, displaying an editing result of the first editing operation and recording the first editing operation in the video producing task; in response to the second editing operation of a second user for the video producing task, displaying an editing result of the second editing operation and recording the second editing operation in the video producing task; and generating a production result of the video producing task based on editing operations recorded in the video producing task.
    Type: Application
    Filed: November 8, 2023
    Publication date: April 10, 2025
    Inventor: Xinwei Li
  • Publication number: 20250119549
    Abstract: During video encoding or decoding, cross-component prediction can be used to predict a chroma sample from collocated reconstructed luma samples. The prediction can be based on a gradient of the collocated reconstructed luma samples, a down-sampled value of the collocated reconstructed luma samples, or a combination thereof. An exemplary method includes: determining a first value associated with a chroma sample, by applying a first gradient pattern to reconstructed values of a first plurality of luma samples; determining a second value associated with the chroma sample, by applying a down-sampling filter to reconstructed values of a second plurality of luma samples; and predicting the chroma sample based on the first value and the second value.
    Type: Application
    Filed: December 19, 2024
    Publication date: April 10, 2025
    Inventors: Xinwei LI, Yan YE, Ru-ling LIAO, Jie CHEN
  • Publication number: 20250113052
    Abstract: Methods are provided for optical flow-based motion refinement. An exemplary method includes: receiving a bitstream; and decoding, using coded information of the bitstream, one or more pictures, wherein the decoding includes: determining whether a current block is coded by partitioning into multiple parts; and in response to the current block being partitioned into multiple parts that are inter predicted according to respective reference blocks, determining whether to apply a local illumination compensation (LIC) process to the multiple parts based on whether a coded block applies the LIC process.
    Type: Application
    Filed: September 24, 2024
    Publication date: April 3, 2025
    Inventors: Xinwei LI, Ru-ling LIAO, Jie CHEN, Yan YE
  • Publication number: 20250113050
    Abstract: Methods and apparatuses are provided for subblock-based temporal motion vector prediction (SbTMVP). An exemplary decoding method includes: decoding a target coding unit (CU) based on a SbTMVP mode, wherein the target CU is divided into a plurality of subblocks; determining a same reference picture index for the plurality of subblocks; selecting, based on the determined reference picture index, a reference picture from a list of reference pictures; generating a plurality of predicted subblocks based on the selected reference picture; and reconstructing the target CU based on the plurality of predicted subblocks.
    Type: Application
    Filed: September 24, 2024
    Publication date: April 3, 2025
    Inventors: Ru-ling LIAO, Yan YE, Jie CHEN, Xinwei LI, Yeping ZHENG
  • Publication number: 20250113053
    Abstract: Methods and apparatuses are provided for motion template-matching-based motion refinement. An exemplary method includes: dividing a target coding block into a plurality of subblocks; determining a first plurality of templates associated with the plurality of subblocks; determining a plurality of reference templates based on the first plurality of templates; performing motion compensation to the target coding block based on an affine merge candidate; and refining the motion compensation by matching the first plurality of templates and the plurality of reference templates.
    Type: Application
    Filed: September 25, 2024
    Publication date: April 3, 2025
    Inventors: Jie CHEN, Ru-Ling LIAO, Xinwei LI, Yan YE
  • Patent number: 12267503
    Abstract: The present disclosure provides a video data processing method for cross-component sample adaptive offset (CCSAO). The method includes determining an index based on a vertical coordinate of a chroma sample within a picture; determining a luma sample based on the index; classifying the chroma sample based on a reconstructed value associated with the luma sample; determining an offset based on the classification; and adding the offset to a reconstructed value associated with the chroma sample.
    Type: Grant
    Filed: November 8, 2023
    Date of Patent: April 1, 2025
    Assignee: Alibaba (China) Co., Ltd.
    Inventors: Xinwei Li, Jie Chen, Ru-Ling Liao, Yan Ye
  • Patent number: 12267493
    Abstract: The present disclosure provides a video data processing method. The method includes: receiving a bitstream; decoding a first index from the bitstream; determining a maximum number of an adaptive loop filter (ALF) for a component of a picture based on the first index; and processing pixels in the picture with the ALF.
    Type: Grant
    Filed: November 16, 2023
    Date of Patent: April 1, 2025
    Assignee: Alibaba Group Holding Limited
    Inventors: Jie Chen, Ru-Ling Liao, Xinwei Li, Yan Ye
  • Patent number: D1079823
    Type: Grant
    Filed: September 25, 2023
    Date of Patent: June 17, 2025
    Inventor: Xinwei Li