| 摘要 |
A set of motion vector predictors is generated for a block to encode forming part of a current video frame that is to be encoded. The number of temporal motion vector predictors in the set is one and the number of spatial motion vector predictors in the set is at least one. The one temporal predictor of the set is determined by selecting as the motion vector at a predetermined position, in the vicinity of a bottom right position of a block of the reference frame collocated with said block to encode, in a block of the reference frame neighbouring the collocated block. The predetermined position may be the top left position in the block diagonally below and to the right of the collocated block. The temporal predictor selected may be determined in dependence on the selected at least one spatial predictor. This can provide coding efficiency improvements and may increase the diversity of the predictor set. |
