FIELD: video encoding.
SUBSTANCE: information is obtained identifying the first and second motion vector predictors for the block with affine coding in the video signal. At least a first motion vector difference and a second motion vector difference are obtained. Information is obtained identifying the accuracy of the first and second motion vector differences from the first set of available accuracy values, the first set of available accuracy values being different from the second set of available accuracy values used for non-affine outer coding. The first checkpoint motion vector is generated based on the first motion vector predictor and the first motion vector difference, and the second checkpoint motion vector is generated based on the second motion vector predictor and the second motion vector difference. A block prediction is generated using an affine motion model, wherein the affine motion model is characterized by at least a first checkpoint motion vector and a second checkpoint motion vector.
EFFECT: increased efficiency of video coding.
25 cl, 25 dwg
