Tile and slice partitioning in video processing
AssigneeAlibaba Group Holding Limited
InventorsJie Chen, Yan Ye, Ru-Ling Liao
Life of this patent
Estimating the term…
Abstract
The present disclosure provides a computer-implemented method for encoding or decoding video. The method includes encoding or decoding, in a plurality of picture parameter sets (PPS) associated with pictures of a coded layer video sequence (CLVS), corresponding first PPS flags indicating whether pictures are allowed to be partitioned into a plurality of tiles or slices. In a first PPS, a corresponding first PPS flag with a first value indicates a first picture of the CLVS is unpartitioned, and in a second PPS, another corresponding first PPS flag with a second value being different from the first value indicates that a second picture of the CLVS is allowed to be partitioned.
Claims18
A method for encoding or decoding video, comprising:
- encoding or decoding, in a plurality of picture parameter sets (PPS) associated with pictures of a coded layer video sequence (CLVS), corresponding first PPS flags indicating whether pictures are allowed to be partitioned into a plurality of tiles or slices, wherein:
- in a first PPS, a corresponding first PPS flag with a first value indicates a first picture of the CLVS is unpartitioned, and in a second PPS, another corresponding first PPS flag with a second value being different from the first value indicates that a second picture of the CLVS is allowed to be partitioned;
- and
- skipping encoding or decoding, in the first PPS, a second PPS flag indicating whether each subpicture of the first picture comprises a single rectangular slice and determining a value of the second PPS flag equal to a third value, wherein the second PPS flag equal to the third value indicates that each subpicture of the first picture comprises a single rectangular slice.
An apparatus, comprising:
- a memory configured to store instructions;
- and
- one or more processors configured to execute the instructions to cause the apparatus to:
- determine whether a picture is partitioned into a plurality of tiles or slices;
- and
- in response to a determination that the picture is allowed to be partitioned into the plurality of tiles or slices, encode or decode, in a picture parameter set (PPS), a first flag associated with a slice mode being applied to the picture referring to the PPS, wherein:
- when a raster-scan slice mode is applied to partition the picture, encode or decode the first flag with a first value;
- or when a rectangular slice mode is applied to partition the picture, encode or decode the first flag with a second value different from the first value;
- encode or decode a first picture parameter specifying the number of provided slice heights for slices in a corresponding tile, the first picture parameter being equal to 0 in response to the corresponding tile containing a single slice;
- and
- encode or decode a second picture parameter specifying a height of a corresponding rectangular slice in the corresponding tile minus one.
A non-transitory computer-readable storage medium storing a bitstream of video data, wherein the bitstream comprises a picture parameter set (PPS) associated with at least one picture of a coded layer video sequence (CLVS), and the bitstream is processed according to a method comprising:
- encoding or decoding, in the PPS, a first PPS flag indicating whether the associated picture is allowed to be partitioned into a plurality of tiles or slices, wherein the first PPS flag being equal to a first value indicates that the associated picture is unpartitioned, or the first PPS flag being equal to a second value different from the first value indicates that the associated picture is allowed to be partitioned;
- and
- when the first PPS flag is equal to the first value, skipping encoding or decoding, in the PPS, a second PPS flag indicating whether each subpicture of the associated picture comprises a single rectangular slice, and determining a value of the second PPS flag equal to a third value, wherein the second PPS flag equal to the third value indicates that each subpicture of the associated picture comprises a single rectangular slice.
Dependencies are inferred from the claim text.