US 10,893,291 B2
Ultimate motion vector expression with adaptive directional information set
Chao-Hsiung Hung, San Diego, CA (US); Chun-Chi Chen, San Diego, CA (US); Wei-Jung Chien, San Diego, CA (US); and Marta Karczewicz, San Diego, CA (US)
Assigned to Qualcomm Incorporated, San Diego, CA (US)
Filed by QUALCOMM Incorporated, San Diego, CA (US)
Filed on Sep. 26, 2019, as Appl. No. 16/584,875.
Claims priority of provisional application 62/738,468, filed on Sep. 28, 2018.
Prior Publication US 2020/0107043 A1, Apr. 2, 2020
Int. Cl. H04N 19/513 (2014.01); H04N 19/56 (2014.01); H04N 19/139 (2014.01); H04N 19/176 (2014.01); H04N 19/105 (2014.01)
CPC H04N 19/521 (2014.11) [H04N 19/105 (2014.11); H04N 19/139 (2014.11); H04N 19/176 (2014.11); H04N 19/56 (2014.11)] 25 Claims
OG exemplary drawing
1. A method of decoding video data, the method comprising:
determining a list of candidates for a current block of the video data from one or more spatial neighboring blocks in a set of spatial neighboring blocks that spatially neighbor the current block of video data;
determining a distance resolution and a direction resolution for a motion vector of at least one of the candidates;
determining, based on data obtained from a bitstream that comprises an encoded representation of the video data, a base candidate index, a direction index and a distance index;
determining a base candidate based on the base candidate index;
determining a direction based on the direction index, the direction index pointing to the direction in a direction table;
determining a distance based on the distance index, the distance index pointing to the distance in a distance table;
adapting one or more of the direction table or the distance table based on the distance resolution or the direction resolution;
determining a motion vector difference (MVD) based on the direction and the distance;
determining a prediction block using the MVD and a motion vector of the base candidate; and
decoding the current block based on the prediction block.