Spatial-temporal motion vector prediction patterns for video coding

摘要

Techniques for creating a motion vector predictor list based on spatially neighboring blocks and collocated blocks are described. A method of coding video data includes, in a first picture having a coding order from left to right, for a first block, creating a first motion vector predictor list, the first in the first motion vector predictor list comprising: Entry is based on motion vector information of a block to the left of the first block, creating the first motion vector predictor list, the first block in the first picture based on the first motion vector predictor list coding, creating a second motion vector predictor list, for a second block, in a second picture having a right-to-left coding order, wherein a first entry in the second motion vector predictor list is creating the second motion vector predictor list based on motion vector information of the right neighboring block for the second block, and coding the second block in the second picture based on the second motion vector predictor list including the steps of