| 摘要 |
The present invention relates to the field of video coding. Provided are a method and device for building a motion vector list for use in motion vector prediction, solving the problem that a serial mode is employed to build the motion vector lists of at least two PUs in a same coding unit, thus increasing parallel processing capabilities. Specifically comprised are: acquiring adjacent blocks on a spatial-domain of a current prediction block, where the current prediction block is located in a current coding unit; determining applicable adjacent blocks of the current prediction block on the basis of the division mode of the current coding unit, where the applicable adjacent blocks are located external of the current coding unit; acquiring motion vector predictors from the applicable adjacent blocks on the basis of the motion vector predictors of the adjacent blocks and according to a predetermined order, and adding the acquired motion vector predictors into the motion vector list. This is applicable in building motion vector lists. |
