US 11,659,198 B2
Method and apparatus of pattern-based motion vector derivation for video coding
Tzu-Der Chuang, Zhubei (TW); Ching-Yeh Chen, Taipei (TW); and Yu-Wen Huang, Taipei (TW)
Assigned to MEDIATEK INC., Hsinchu (TW)
Filed by MEDIATEK INC., Hsinchu (TW)
Filed on Apr. 9, 2021, as Appl. No. 17/226,354.
Application 17/226,354 is a continuation of application No. 16/083,645, abandoned, previously published as PCT/CN2017/076622, filed on Mar. 14, 2017.
Claims priority of provisional application 62/309,001, filed on Mar. 16, 2016.
Prior Publication US 2021/0227248 A1, Jul. 22, 2021
Int. Cl. H04N 19/52 (2014.01); H04N 19/56 (2014.01); H04N 19/577 (2014.01); H04N 19/44 (2014.01); H04N 19/176 (2014.01); H04N 19/184 (2014.01)
CPC H04N 19/52 (2014.11) [H04N 19/176 (2014.11); H04N 19/184 (2014.11); H04N 19/44 (2014.11); H04N 19/56 (2014.11); H04N 19/577 (2014.11)] 12 Claims
OG exemplary drawing
 
1. A method of video coding using motion compensation, the method comprising:
receiving input data associated with a current block in a current picture, wherein each current block is divided into multiple sub-blocks;
deriving a first-stage motion vector (MV) or a first-stage MV pair using bilateral matching, template matching or both based on one or more first-stage initial MV candidates or one or more first-stage initial MV pair candidates within a first-stage search window;
deriving second-stage MVs for the multiple sub-blocks by deriving one or more second-stage MVs for each sub-block using the bilateral matching, the template matching or both, wherein the first-stage MV or the first-stage MV pair is used as an initial MV or MV pair for second-stage bilateral matching, template matching or both within a second-stage search window, and wherein the second-stage search window is within the first-stage search window, wherein all samples of the second-stage search window are within samples of the first-stage window regardless of a proximity of the first-stage MV or the first-stage MV pair to a boundary of the first-stage search window;
determining final MVs or final motion vector predictors (MVPs) from a set of MV candidates or MVPs candidates including the second-stage MVs; and
encoding or decoding the current block or a current MV of the current block using the final MVs or the final motion vector predictors (MVPs) at an encoder side or a decoder side respectively.