US 11,943,467 B2
Affine motion estimation for affine model-based video coding
Yuwen He, San Diego, CA (US); Xiaoyu Xiu, San Diego, CA (US); Yan Ye, San Diego, CA (US); and Jiancong Luo, Skillman, NJ (US)
Assigned to VID Scale, Inc., Wilmington, DE (US)
Appl. No. 17/278,202
Filed by VID SCALE, INC., Wilmington, DE (US)
PCT Filed Sep. 20, 2019, PCT No. PCT/US2019/052101
§ 371(c)(1), (2) Date Mar. 19, 2021,
PCT Pub. No. WO2020/061423, PCT Pub. Date Mar. 26, 2020.
Claims priority of provisional application 62/734,783, filed on Sep. 21, 2018.
Claims priority of provisional application 62/812,979, filed on Mar. 2, 2019.
Prior Publication US 2021/0400300 A1, Dec. 23, 2021
Prior Publication US 2023/0037273 A2, Feb. 2, 2023
Int. Cl. H04N 19/52 (2014.01); H04N 19/147 (2014.01); H04N 19/176 (2014.01)
CPC H04N 19/52 (2014.11) [H04N 19/147 (2014.11); H04N 19/176 (2014.11)] 20 Claims
OG exemplary drawing
 
1. A video encoding device comprising:
a processor configured to:
during a first partition type check, partition a parent block using a first tree partition to obtain a first coding block;
perform affine motion estimation associated with the first tree partition to derive, for the first coding block, a first motion vector (MV) set;
add the first MV set to a recently-estimated MV list, wherein the recently-estimated MV list comprises a plurality of MV sets derived by performing affine motion estimation associated with the parent block;
during a second partition type check, partition the parent block using a second tree partition to obtain a second coding block;
select, for the second coding block, a starting point for affine motion estimation based at least in part on an MV set in the recently-estimated MV list and an MV predictor; and
perform affine motion estimation associated with the second tree partition to derive a second MV set for the second coding block based on the selected starting point.