FIELD: physics, computer engineering.
SUBSTANCE: invention relates to computer engineering. A method of generating a set of motion vector predictors for a block to be decoded, said block forming part of a current video frame to be decoded, wherein the number of temporal motion vector predictors in the set is equal to one and the number of spatial motion vector predictors in the set is at least equal to one, the method comprising a step of determining said one temporal predictor from the set by selecting as the one temporal predictor the motion vector at a predetermined position in a block of the reference frame neighbouring said block of the reference frame located identically with said block to be decoded, wherein said neighbouring block is located lower on a diagonal and to the right of the block of the reference frame which is located identically with said block to be decoded, and said predetermined position is located in the top left position in said neighbouring block.
EFFECT: reduced memory for storing motion vectors.
7 cl, 15 dwg, 2 tbl
