Unified design for picture partitioning schemes

摘要

A video coder can control in-picture prediction across slice boundaries within a picture. In one example, a first syntax element can control if in-picture prediction across slice boundaries for slices of a picture. If in-picture prediction across slice boundaries is enabled for the picture, then a second syntax element can control, for an individual slices, if in-picture prediction across slice boundaries is enabled for the slice.

主权项

1. A method of coding video data, the method comprising:
coding, in a picture parameter set, a first syntax element for a first picture of the video data, wherein a first value for the first syntax element indicates that in-picture prediction is allowed across slices for slices of the first picture by indicating that syntax elements in headers for one or more slices of the first picture are allowed to be predicted from syntax elements in a header of a preceding slice of the first picture; in response to the first syntax element indicating that in-picture prediction is allowed across slices, coding, in respective slice headers of slices in the first picture, a second syntax element indicating whether in-picture prediction is allowed across that slice by indicating whether syntax elements in a header of the respective slice are predicted from syntax elements in the header of the preceding slice, wherein presence of the second syntax element in the slice header is dependent on the first value of the first syntax element indicating that in-picture prediction is allowed across slices for slices of the first picture; and, in response to the first syntax element indicating that in-picture prediction is allowed across slices and a respective second syntax element, for a first slice of the first picture, indicating that in-picture prediction is allowed across the first slice, coding a first coding unit of the first slice of the first picture of the video data based on information of a second coding unit of a second slice of the first picture of the video data.