CPC H04N 19/122 (2014.11) [H04N 19/176 (2014.11); H04N 19/523 (2014.11); H04N 19/547 (2014.11)] | 22 Claims |
1. A method, comprising:
performing a first sub-block-based affine transform prediction on an affine block when a first sub-block size of a sub-block of the affine block is greater than or equal to a first sub-block size threshold;
detecting, according to values of control point motion vectors (CPMVs) for the affine block, whether a pixel-based motion vector field is not applied for performing motion compensation for the affine block when the first sub-block size is less than the first sub-block size threshold;
when detecting that the pixel-based motion vector field is not applied, performing a second sub-block-based affine transform prediction on the affine block with a second sub-block size and not performing the motion compensation for the affine block using the pixel-based motion vector field, wherein the second sub-block size is based on the first sub-block size; and
when detecting that the pixel-based motion vector field is applied, performing the motion compensation for the affine block using the pixel-based motion vector field.
|