Sub-pictures for pixel rate balancing on multi-core platforms

1. A method for decoding a compressed bit stream in a video decoder to recover a video sequence, the video decoder comprising a plurality of decoder processing cores, the method comprising:

determining, by the video decoder, that a picture is encoded in the compressed bit stream in a predetermined number of independently encoded sub-pictures, each of the sub-pictures completely containing one or more complete slices, each of the sub-pictures completely containing one or more complete tiles, wherein a number of tile rows per picture is constrained to be equal to a number of sub-picture rows per picture, and wherein each sub-picture comprises an integer number of coding units in a rectangle; and

dispatching, by the video decoder, a first encoded sub-picture of the predetermined number of sub-pictures to a first decoder processing core of the plurality of decoder processing cores and a second encoded sub-picture of the predetermined number of sub-pictures to a second decoder processing core of the plurality of decoder processing cores, wherein the first encoded sub-picture and the second encoded sub-picture are independently decoded in parallel on the respective first and second decoder processing cores.

2. The method of claim 1 , wherein a size difference between a number of rows of LCUs in a largest sub-picture and a number of rows of coding units in a smallest sub-picture is less than or equal to one coding unit, and wherein a size difference between a number of columns of coding units in the largest sub-picture and a number of columns of coding units in the smallest sub-picture is less than or equal to one coding unit.

3. The method of claim 1 , wherein sub-picture partitioning of the picture is signaled in the compressed bit stream using a tile signaling protocol.

4. The method of claim 1 , wherein dispatching further comprises:

determining an entry point of a first sub-picture bit stream in the compressed bit stream, the first sub-picture bit stream corresponding to a first sub-picture of the sub-pictures;

determining an entry point of a second sub-picture bit stream in the compressed bit stream, the second sub-picture bit stream corresponding to a second sub-picture of the sub-pictures;

directing the first sub-picture bit stream to the first decoder processing core; and

directing the second sub-picture bit stream to the second decoder processing core.

5. The method of claim 4 , wherein the entry points of the first and second sub-picture bit streams are signaled in the compressed bit stream.

6. The method of claim 4 , wherein dispatching further comprises:

determining that a first slice encoded in the compressed bit stream is a slice corresponding to the first sub-picture;

directing a slice bit stream corresponding to the first slice to the first decoder processing core;

determining that a second slice encoded in the compressed bit stream is a slice corresponding to the second sub-picture; and

directing a slice bit stream corresponding to the second slice to the second decoder processing core,

wherein slice bit streams corresponding to the first sub-picture and slice bit streams corresponding to the second sub-picture are interleaved in the compressed bit stream.

7. The method of claim 1 , wherein the predetermined number of sub-pictures is mandated by a video coding standard.

8. The method of claim 7 , wherein the predetermined number of sub-pictures is specified as part of a level defined by the video coding standard.

9. The method of claim 7 , wherein the predetermined number of sub-pictures is specified as a required number of sub-picture rows and a required number of sub-picture columns.

10. The method of claim 7 , wherein the predetermined number of sub-pictures is specified as a required number of sub-pictures and a minimum number of sub-picture columns.

11. A method for encoding a digital video sequence in a video encoder to generate a compressed bit stream, the video encoder comprising a plurality of encoder processing cores, the method comprising:

partitioning, by the video encoder, a picture into a predetermined number of sub-pictures, wherein each of the sub-pictures completely contains one or more complete slices, wherein each sub-picture can be independently encoded, wherein each of the sub-pictures completely contains one or more complete tiles, and wherein a number of tile rows per picture is constrained to be equal to a number of sub-picture rows per picture, and wherein each sub-picture comprises an integer number of coding units in a rectangle;

dispatching, by the video encoder, a first sub-picture of the predetermined number of sub-pictures to a first encoder processing core of the plurality of encoder processing cores and dispatching a second sub-picture of the predetermined number of sub-pictures to a second encoder processing core of the plurality of encoder processing cores; and

encoding, by the video encoder, the first sub-picture and the second sub-picture in parallel on the respective first and second encoder processing core.

12. The method of claim 11 , wherein the predetermined number of sub-pictures is mandated by a video coding standard.

13. The method of claim 12 , wherein the predetermined number of sub-pictures is specified as part of a level defined by the video coding standard.

14. The method of claim 12 , wherein the predetermined number of sub-pictures is specified as a required number of sub-picture rows and a required number of sub-picture columns.

15. The method of claim 12 , wherein the predetermined number of sub-pictures is specified as a required number of sub-pictures and a minimum number of sub-picture columns.

16. The method of claim 11 , wherein a size difference between a number of rows of coding units in a largest sub-picture and a number of rows of coding units in a smallest sub-picture is less than or equal to one coding unit, and wherein a size difference between a number of columns of coding units in the largest sub-picture and a number of columns of coding units in the smallest sub-picture is less than or equal to one coding unit.

17. The method of claim 11 , wherein sub-picture partitioning of the picture is signaled in the compressed bit stream using a tile signaling protocol.

18. The method of claim 11 , further comprising:

signaling entry points of a first sub-picture bit stream and a second sub-picture bit stream in the compressed bit stream, wherein encoding the first sub-picture on the first encoder processing core generates the first sub-picture bit stream and encoding the second sub-picture on the second encoder processing core generates the second sub-picture bit stream.

19. The method of claim 18 , further comprising:

assigning a first sub-picture identifier to the first sub-picture and a second sub-picture identifier to the second sub-picture, and

wherein encoding the first sub-picture and the second sub-picture further comprises:

generating a first plurality of slice bit streams for the first sub-picture;

generating a second plurality of slice bit streams for the second sub-picture;

signaling the first sub-picture identifier in the compressed bit stream in association with each slice bit stream of the first plurality of slice bit streams; and

signaling the second sub-picture identifier in the compressed bit stream in association with each slice bit stream of the second plurality of slice bit streams,

wherein the first plurality of slice bit streams and the second plurality of slice bit streams are interleaved in the compressed bit stream.