US 12,425,636 B2
Segmentation-based parameterized motion models
Debargha Mukherjee, Cupertino, CA (US); Yuxin Liu, Palo Alto, CA (US); and Sarah Parker, San Francisco, CA (US)
Assigned to GOOGLE LLC, Mountain View, CA (US)
Filed by GOOGLE LLC, Mountain View, CA (US)
Filed on Nov. 28, 2023, as Appl. No. 18/521,182.
Application 18/521,182 is a continuation of application No. 16/693,425, filed on Nov. 25, 2019, abandoned.
Application 16/693,425 is a continuation of application No. 15/838,748, filed on Dec. 12, 2017, granted, now 10,506,249, issued on Dec. 10, 2019.
Claims priority of provisional application 62/471,659, filed on Mar. 15, 2017.
Prior Publication US 2024/0098298 A1, Mar. 21, 2024
Int. Cl. H01L 29/94 (2006.01); H04N 19/17 (2014.01); H04N 19/20 (2014.01); H04N 19/513 (2014.01); H04N 19/517 (2014.01); H04N 19/54 (2014.01); H04N 19/543 (2014.01); H04N 19/547 (2014.01); H04N 19/557 (2014.01); H04N 19/80 (2014.01)
CPC H04N 19/517 (2014.11) [H04N 19/17 (2014.11); H04N 19/20 (2014.11); H04N 19/521 (2014.11); H04N 19/54 (2014.11); H04N 19/543 (2014.11); H04N 19/547 (2014.11); H04N 19/557 (2014.11); H04N 19/80 (2014.11)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
decoding, from a compressed bitstream and with respect to each reference frame of two or more reference frames available for decoding a current frame, at least two global motion models, wherein decoding a global motion model comprises decoding parameters for the global motion model, wherein the global motion models constitute multiple global motion models and wherein each global motion model is based on a segmentation of the current frame and represents a respective underlying motion of blocks within a respective segment; and
decoding blocks of the current frame by:
for each inter-predicted block of a segment, decoding, from the compressed bitstream, a per block indication specific to the each inter-predicted block indicating either
to generate a prediction block for the each inter-predicted block based on a global motion model of the multiple global motion models and associated with one of the two or more reference frames, or
to generate the prediction block for the each inter-predicted block based on a motion vector that is different from the global motion model; and
decoding the each inter-predicted block based on the per block indication.