US 11,877,004 B2
Constraining motion vectors of groups of blocks
Xiang Li, Saratoga, CA (US); Meng Xu, San Jose, CA (US); Xiaozhong Xu, State College, PA (US); Shan Liu, San Jose, CA (US); and Guichun Li, San Jose, CA (US)
Assigned to TENCENT AMERICA LLC, Palo Alto, CA (US)
Filed by TENCENT AMERICA LLC, Palo Alto, CA (US)
Filed on Apr. 7, 2023, as Appl. No. 18/297,442.
Application 18/297,442 is a continuation of application No. 17/458,680, filed on Aug. 27, 2021, granted, now 11,671,621.
Application 17/458,680 is a continuation of application No. 16/517,156, filed on Jul. 19, 2019, granted, now 11,184,635, issued on Nov. 23, 2021.
Claims priority of provisional application 62/788,838, filed on Jan. 5, 2019.
Claims priority of provisional application 62/734,987, filed on Sep. 21, 2018.
Claims priority of provisional application 62/725,927, filed on Aug. 31, 2018.
Prior Publication US 2023/0247218 A1, Aug. 3, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. H04N 19/513 (2014.01); H04N 19/44 (2014.01)
CPC H04N 19/521 (2014.11) [H04N 19/44 (2014.11)] 20 Claims
OG exemplary drawing
 
1. A method for video encoding in an encoder, comprising:
obtaining a current block having a size of W×H pixels and including a plurality of sub-blocks according to a sub-block based inter prediction mode, the plurality of sub-blocks being arranged into one or more areas having a size of M×N pixels in the current block, M≤W, and N≤H;
for each area of the one or more areas having the size of M×N pixels in the current block:
determining base motion vectors of a portion of the sub-blocks within a corresponding area, the portion including at least one non-corner sub-block of the current block;
converting the base motion vectors of the portion of the sub-blocks into respective target motion vectors of the portion of the sub-blocks, the respective target motion vectors being confined by a first lower bound and a first upper bound along a first coordinate direction and by a second lower bound and a second upper bound along a second coordinate direction, and the converting comprising, for a base motion vector of the at least one non-corner sub-block in the portion of the sub-blocks within the corresponding area, clipping values exceeding one of the first lower bound, the first upper bound, the second lower bound, or the second upper bound; and
encoding samples of the corresponding area according to the target motion vectors of the portion of the sub-blocks; and
generating a coded video bitstream including (i) the encoded samples of each of the one or more areas and (ii) prediction information of the current block, the prediction information indicating the sub block based inter prediction mode and indicating the base motion vectors of the portion of the subblocks.