Search

EP-4736412-A1 - MERGE MODE INTRA-TEMPLATE MATCHING PREDICTION

EP4736412A1EP 4736412 A1EP4736412 A1EP 4736412A1EP-4736412-A1

Abstract

For encoding and decoding video with intra-template matching prediction, a merge mode may be used. A device for video decoding and/or encoding may identify a first block vector-predicted video block that neighbors a second video block. The first block vector-predicted video block may be associated with intra-template matching prediction (intraTMP). The device may obtain an intra-template parameter for the first block vector-predicted video block. The device may infer a candidate intraTMP parameter for the second video block based on the intra-template parameter for the first block vector-predicted video block. The device may decode and/or encode the second video block based on the candidate intraTMP parameter. The intra-template parameters may include parameters such as an intra-template fusion flag, an intra-template fusion index, an intra-template fusion weight type, an intra-template index, an intra-template fusion flag, an intra-template sub-pel precision index, an intra-template sub-pel direction index, and/or the like.

Inventors

  • NASER, Karam
  • LE LEANNEC, FABRICE
  • ROBERT, ANTOINE
  • GALPIN, FRANCK

Assignees

  • InterDigital CE Patent Holdings, SAS

Dates

Publication Date
20260506
Application Date
20240606

Claims (18)

  1. 1 . A device for video decoding comprising a processor, wherein the processor is configured to: identify a first block vector-predicted video block that neighbors a second video block, wherein the first block vector-predicted video block is associated with intra-template matching prediction (intraTM P); obtain an intra-template parameter for the first block vector-predicted video block; infer a candidate intraTM P parameter for the second video block based on the intra-template parameter for the first block vector-predicted video block; and decode the second video block based on the candidate intraTMP parameter.
  2. 2. A device for video encoding comprising a processor, wherein the processor is configured to: identify a first block vector-predicted video block that neighbors a second video block, wherein the first block vector-predicted video block is associated with intra-template matching prediction (intraTMP); obtain an intra-template parameter for the first block vector-predicted video block; determine a candidate intraTMP parameter for the second video block based on the intra-template parameter for the first block vector-predicted video block; and encode the second video block based on the candidate intraTMP parameter.
  3. 3. The device of claim 1 , wherein the processor is further configured to: receive, in video data, an indication indicating that intra-template merge mode is to be used for the second video block, wherein the inference of the candidate intraTMP parameter for the second video block is performed based on the indication indicating that intra-template merge mode is to be used for the second video block.
  4. 4. The device of claim 2, wherein the processor is further configured to: indicate, in video data, that intra-template merge mode is to be used for the second video block.
  5. 5. The device of any one of claims 1 -4, wherein the intra-template parameter for the first block vector- predicted video block comprises at least one of an intra-template fusion flag, an intra-template fusion index, an intra-template fusion weight type, an intra-template index, an intra-template sub-pel precision index, or an intra-template sub-pel direction index.
  6. 6. The device of any one of claims 1 -5, wherein the intra-template parameter for the first block vector- predicted video block comprises a block vector associated with the first video block.
  7. 7. The device of any one of claims 1 -6, wherein the processor is further configured to: refine the candidate intraTMP parameter based on a template of the second video block.
  8. 8. The device of any one of claims 1 -5, wherein the intra-template parameter for the first block vector- predicted video block comprises a block vector associated with the first video block, and the processor is further configured to: identify a reduced search area based on the block vector associated with the first video block; and identify a refined block vector for the second block in the reduced search area by minimizing a template cost.
  9. 9. A video decoding method comprising: identifying a first block vector-predicted video block that neighbors a second video block, wherein the first block vector-predicted video block is associated with intra-template matching prediction (intraTMP); obtaining an intra-template parameter for the first block vector-predicted video block; inferring a candidate intraTMP parameter for the second video block based on the intra-template parameter for the first block vector-predicted video block; and decoding the second video block based on the candidate intraTMP parameter.
  10. 10. A video encoding method comprising: identifying a first block vector-predicted video block that neighbors a second video block, wherein the first block vector-predicted video block is associated with intra-template matching prediction (intraTMP); obtaining an intra-template parameter for the first block vector-predicted video block; determining a candidate intraTMP parameter for the second video block based on the intra- template parameter for the first block vector-predicted video block; and encoding the second video block based on the candidate intraTMP parameter.
  11. 11 . The method of claim 9, wherein the method further comprises: receiving, in video data, an indication indicating that intra-template merge mode is to be used for the second video block, wherein the inference of the candidate intraTMP parameter for the second video block is performed based on the indication indicating that intra-template merge mode is to be used for the second video block.
  12. 12. The method of claim 10, wherein the method further comprises: indicating, in video data, that intra-template merge mode is to be used for the second video block.
  13. 13. The method of any one of claims 9-12, wherein the intra-template parameter for the first block vector-predicted video block comprises at least one of an intra-template fusion flag, an intra-template fusion index, an intra-template fusion weight type, an intra-template index, an intra-template sub-pel precision index, or an intra-template sub-pel direction index.
  14. 14. The method of any one of claims 9-13, wherein the intra-template parameter for the first block vector-predicted video block comprises a block vector associated with the first video block.
  15. 15. The method of any one of claims 9-14, further comprising: refining the candidate intraTMP parameter based on a template of the second video block.
  16. 16. The method of any one of claims 9-13, wherein the intra-template parameter for the first vector- predicted video block comprises a block vector associated with the first video block, and wherein the method further comprises: identifying a reduced search area based on the block vector associated with the first video block; and identifying a refined block vector for the second block in the reduced search area by minimizing a template cost.
  17. 17. A computer program product which is stored on a non-transitory computer readable medium and comprises program code instructions for implementing the steps of a method according to claim 10, 12, or claim 10 and any one of claims 13-16, when executed by a processor.
  18. 18. Video data comprising information representative of the encoded current block generated according to the method of claim 10, 12, or claim 10 and any one of claims 13-16.

Description

MERGE MODE INTRA-TEMPLATE MATCHING PREDICTION CROSS-REFERENCE TO RELATED APPLICATIONS [0001] This application claims the benefit of European Provisional Patent Application No. 23306109.2, filed June 30, 2023, the contents of which are hereby incorporated by reference herein. BACKGROUND [0002] Video coding systems may be used to compress digital video signals, e.g., to reduce the storage and/or transmission bandwidth needed for such signals. Video coding systems may include, for example, block-based, wavelet-based, and/or object-based systems. SUMMARY [0003] For encoding and decoding video with intra-template matching prediction, a merge mode may be used. Intra-template parameters used for decoding a coding unit that is intra-template matching prediction encoded may be inherited for use decoding a neighboring coding unit. Signaling may include information indicating that the intra-template parameters for a coding unit may be inherited for use with a neighboring coding unit. The intra-template parameters may include parameters such as an intra-template fusion flag, an intra-template fusion index, an intra-template fusion weight type, an intra-template index, an intra- template fusion flag, an intra-template sub-pel precision index, an intra-template sub-pel direction index, or the like. [0004] In examples, a device for video decoding and/or encoding may identify a first block vector- predicted video block that neighbors a second video block. The first block vector-predicted video block may be associated with intra-template matching prediction (intraTM P). The device may obtain an intra-template parameter for the first block vector-predicted video block. The device may infer a candidate intraTM P parameter for the second video block based on the intra-template parameter for the first block vector- predicted video block. The device may decode and/or encode the second video block based on the candidate intraTM P parameter. [0005] The device may include one or more features as described herein. The device may receive, in video data, an indication indicating that intra-template merge mode may be used for the second video block. The inference of the candidate intraTMP parameter for the second video block may be performed based on the indication indicating that intra-template merge mode may be used for the second video block. The device may indicate, in video data, that intra-template merge mode is to be used for the second video block. The intra-template parameter for the first block vector-predicted video block may be at least one of an intra-template fusion flag, an intra-template fusion index, an intra-template fusion weight type, an intra- template index, an intra-template sub-pel precision index, or an intra-template sub-pel direction index. The intra-template parameter for the first block vector-predicted video block may include a block vector associated with the first video block. The intra-template parameter for the first block vector-predicted video block may include a block vector associated with the first video block. The device may be configured to identify a reduced search area based on the block vector associated with the first video block. The device may identify a refined block vector for the second block in the reduced search area by minimizing a template cost. [0006] A video encoding and/or decoding method may include identifying a first block vector-predicted video block that neighbors a second video block. The first block vector-predicted video block may be associated with intra-template matching prediction (intraTMP). The method may include obtaining an intra- template parameter for the first block vector-predicted video block. The method may include determining a candidate intraTMP parameter for the second video block based on the intra-template parameter for the first block vector-predicted video block. The method may include encoding and/or deciding the second video block based on the candidate intraTMP parameter. [0007] The method may include one or more features as described herein. The method may include receiving, in video data, an indication indicating that intra-template merge mode may be used for the second video block. The inference of the candidate intraTMP parameter for the second video block may be performed based on the indication indicating that intra-template merge mode may be used for the second video block. The method may include indicating, in video data, that intra-template merge mode is to be used for the second video block. The intra-template parameter for the first block vector-predicted video block may be at least one of an intra-template fusion flag, an intra-template fusion index, an intra-template fusion weight type, an intra-template index, an intra-template sub-pel precision index, or an intra-template sub-pel direction index. The intra-template parameter for the first block vector-predicted video block may include a block vector associated with the first video b