US 12,309,413 B2
Efficient affine merge motion vector derivation
Kai Zhang, San Diego, CA (US); Li Zhang, San Diego, CA (US); Hongbin Liu, Beijing (CN); and Yue Wang, Beijing (CN)
Assigned to BEJING BYTEDANCE NETWORK TECHNOLOGY CO., LTD., Beijing (CN); and BYTEDANCE INC., Los Angeles, CA (US)
Filed by Beijing Bytedance Network Technology Co., Ltd., Beijing (CN); and Bytedance Inc., Los Angeles, CA (US)
Filed on Sep. 30, 2021, as Appl. No. 17/490,220.
Application 17/490,220 is a continuation of application No. 17/090,122, filed on Nov. 5, 2020, abandoned.
Application 17/090,122 is a continuation of application No. PCT/IB2019/055592, filed on Jul. 1, 2019.
Claims priority of application No. PCT/CN2018/093943 (WO), filed on Jul. 1, 2018; and application No. PCT/CN2018/095568 (WO), filed on Jul. 13, 2018.
Prior Publication US 2022/0046267 A1, Feb. 10, 2022
Int. Cl. H04N 19/513 (2014.01); H04N 19/176 (2014.01)
CPC H04N 19/513 (2014.11) [H04N 19/176 (2014.11)] 14 Claims
OG exemplary drawing
 
1. A method of processing video data, comprising:
determining, for a conversion between a current video block of a video and a bitstream of the video, motion vectors at control points (CPMVs) of the current video block based on a rule; and
performing the conversion between the current video block and the bitstream based on the CPMVs,
wherein performing the conversion comprises:
determining a motion vector for each sub-block of a multiple sub-blocks of the current video block based on the CPMVs and a specific position of a corresponding sub-block;
rounding the motion vector of each subblock to 1/16 fraction accuracy; and
filtering the rounded motion vector of each subblock with a filter to generate a prediction of each subblock,
wherein a maximum accuracy of the filter is 1/16 fractional pel,
wherein the rule specifies that a CPMV candidate of the current video block is derived from motion vectors from top adjacent blocks coded with an affine mode, and the CPMV candidate is used to derive the CPMVs of the current video block, and
wherein the current video block belongs to a current slice, and a spatial neighboring block of the current video block is invalid for determining CPMVs of the current video block in a case that the spatial neighboring block belongs to a slice different from the current slice.