# US 12,604,000 B2 · Systems and methods for partition-based prediction mode reordering > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,604,000 B2 - Document ID: US12604000B2 - Status: Granted - Kind: B2 - Application number: 18/244,837 - Filed: 2023-09-11 - Published: 2024-08-15 - Granted: 2026-04-14 - Earliest claimed: 2023-02-03 - Assignees: Tencent America LLC - Applicants: Tencent America LLC - Inventors: Han Gao; Xin Zhao; Liang Zhao; Jing Ye; Shan Liu - Web page: https://patentgrep.com/patent/US12604000B2 ## CPC - H04N 19/119: Adaptive subdivision aspects, e.g. subdivision of a picture into rectangular or non-rectangular coding blocks - H04N 19/105: Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction - H04N 19/132: Sampling, masking or truncation of coding units, e.g. adaptive resampling, frame skipping, frame interpolation or high-frequency transform coefficient masking - H04N 19/176: the region being a block, e.g. a macroblock - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION ## Abstract The various implementations described herein include methods and systems for encoding and decoding video. In one aspect, a method of video decoding includes receiving video data that includes a first block from a video bitstream. The method also includes obtaining a set of partition modes and generating a reference template for the first block. The method further includes identifying, from the video bitstream, an index for the reordered set of partition modes, and reconstructing the first block using a first partition mode of the reordered set of partition modes, where the first partition mode corresponds to the index. ## Claims ### Claim 1 A method of video decoding performed at a computing system having memory and one or more processors, the method comprising: receiving video data comprising a plurality of blocks, including a first block and a reference block associated with the first block, from a video bitstream, wherein the first block comprises at least two partitions and is to be predicted via a partition-based prediction; obtaining a set of partition modes for partitioning the first block, each partition mode being uniquely associated with a partition line that partitions the first block into the at least two partitions, wherein each partition mode in the set of partition modes corresponds to a respective partition angle of a set of partition angles and a respective partition offset of a set of partition offsets; generating both a reference template using samples of the reference block and a reconstruction template using samples of the first block, wherein the reference template is generated using integer samples of the reference block and blended using additional fractional samples of the reference block; calculating, for each partition mode of the set of partition modes, a cost between the reference template and the reconstruction template; reordering the set of partition angles corresponding to the set of partition modes based on the calculated cost between the reference template and the reconstruction template for each partition mode; obtaining, from the video bitstream, an index value indicating a partition angle from the reordered set of partition angles; selecting a partition mode according to the index value indicating the partition angle from the reordered set of partition angles and a partition offset of the set of partition offsets; and reconstructing the first block using the selected partition mode. ### Claim 2 (depends on 1) The method of claim 1, wherein the reference template is generated using top and left lines of integer position samples of the reference block, and wherein the reconstruction template generated is using top and left lines of samples of the first block. ### Claim 3 (depends on 1) The method of claim 1, wherein each partition mode in the set of partition modes corresponds to a respective partition sign of a set of partition signs, wherein a first partition sign of the set of partition signs is signaled in the video bitstream, and wherein the set of partition angles are reordered based on the first partition sign. ### Claim 4 (depends on 1) The method of claim 1, further comprising reordering the set of partition offsets and selecting a partition offset from the reordered set of partition offsets. ### Claim 5 (depends on 1) The method of claim 1: wherein the set of partition modes is grouped into a plurality of groups; the method further comprises a group index of a selected group of the plurality of groups. ### Claim 6 (depends on 1) The method of claim 1, wherein the index value is coded using multi-symbol entropy coding. ### Claim 7 (depends on 1) The method of claim 1, wherein a maximum value for the index value is predefined or signaled in high-level syntax. ### Claim 8 A method of video encoding performed at a computing system having memory and one or more processors, the method comprising: receiving video data comprising a plurality of blocks, including a first block and a reference block associated with the first block, wherein the first block comprises at least two partitions and is to be predicted via a partition-based prediction; obtaining a set of partition modes for partitioning the first block, each partition mode being uniquely associated with a partition line that partitions the first block into the at least two partitions, wherein each partition mode in the set of partition modes corresponds to a respective partition angle of a set of partition angles and a respective partition offset of a set of partition offsets; generating both a reference template using samples of the reference block and a reconstruction template using samples of the first block, wherein the reference template is generated using integer samples of the reference block and blended using additional fractional samples of the reference block; calculating, for each partition mode of the set of partition modes, a cost between the reference template and the reconstruction template; reordering the set of partition angles corresponding to the set of partition modes based on the calculated cost between the reference template and the reconstruction template for each partition mode; selecting a partition angle from the reordered set of partition angles; signaling, via a video bitstream, an index value indicating the selected partition angle; and encoding the first block according to the selected partition angle and a partition offset of the set of partition offsets. ### Claim 9 (depends on 8) The method of claim 8, wherein the reference template is generated using top and left lines of integer position samples of the reference block, and wherein the reconstruction template generated is using top and left lines of samples of the first block. ### Claim 10 (depends on 8) The method of claim 8, wherein the reordered set of partition modes is a subset of the set of partition modes. ### Claim 11 (depends on 8) The method of claim 8, wherein each partition mode in the set of partition modes corresponds to a respective partition sign of a set of partition signs, wherein a first partition sign of the set of partition signs is signaled in the video bitstream, and wherein the set of partition angles are reordered based on the first partition sign. ### Claim 12 (depends on 8) The method of claim 8, further comprising reordering the set of partition offsets and selecting a partition offset from the reordered set of partition offsets. ### Claim 13 (depends on 8) The method of claim 8: wherein the set of partition modes is grouped into a plurality of groups; the method further comprises signaling a group index of a selected group of the plurality of groups. ### Claim 14 (depends on 8) The method of claim 8, further comprising encoding the index value using multi-symbol entropy coding. ### Claim 15 (depends on 8) The method of claim 8, further comprising signaling a maximum value for the index value in high-level syntax of the video bitstream. ### Claim 16 A method for obtaining a video bitstream, the video bitstream comprising coded information for a plurality of blocks of video data, the plurality of blocks including a first block, and an indicator indicating an index value indicating a partition angle for the first block, the method comprising: obtaining a set of partition modes for partitioning the first block, each partition mode being uniquely associated with a partition line that partitions the first block into at least two partitions, wherein each partition mode in the set of partition modes corresponds to a respective partition angle of a set of partition angles and a respective partition offset of a set of partition offsets; generating both a reference template using samples of the reference block and a reconstruction template using samples of the first block, wherein the reference template is generated using integer samples of the reference block and blended using additional fractional samples of the reference block; calculating, for each partition mode of the set of partition modes, a cost between the reference template and the reconstruction template; reordering the set of partition angles corresponding to the set of partition modes based on the calculated cost between the reference template and the reconstruction template for each partition mode; selecting a partition angle from the reordered set of partition angles; signaling the index value indicating the selected partition angle; and encoding the first block according to the selected partition angle and a partition offset of the set of partition offsets. ### Claim 17 (depends on 16) The method of claim 16, wherein the reference template is generated using top and left lines of integer position samples of the reference block, and wherein the reconstruction template generated is using top and left lines of samples of the first block. ### Claim 18 (depends on 16) The method of claim 16, wherein the video bitstream further comprises a second indicator indicating a partition sign for the selected partition angle. ### Claim 19 (depends on 16) The method of claim 16, wherein the indicator indicating the index value is encoded using multi-symbol entropy coding. ### Claim 20 (depends on 16) The method of claim 16, wherein the video bitstream further comprises a second indicator indicating a maximum value for the index value.