Hardware optimization of rate calculation in rate distortion optimization for video encoding

1. A method, comprising:

for each candidate prediction mode among different prediction modes of a video codec:

selecting by an application-specific integrated circuit one of a plurality of different probability lookup tables that corresponds to the candidate prediction mode;

using by the application-specific integrated circuit the selected one of the plurality of different probability lookup tables to calculate a corresponding token rate for the candidate prediction mode for a video to be encoded; and

using by the application-specific integrated circuit the corresponding token rate for the candidate prediction mode to determine a rate distortion cost for the video; and

encoding the video using a selected one of the different prediction modes determined based on the determined rate distortion costs.

2. The method of claim 1 , wherein a prediction mode comprises one or more parameters defined by a standardized video coding format, wherein the one or more parameters comprise a transform unit size, and wherein the standardized video coding format is selected from a group consisting of: H.262, H.264, H.265, Theora, RealVideo RV40, VP9, and AV1.

3. The method of claim 1 , wherein each of the plurality of different probability lookup tables comprises a partition of an original probability lookup table that is partitioned based on one or more parameters defined by a video coding standard.

4. The method of claim 3 , wherein the partitions of the original probability lookup table are partitioned based on a type of prediction that exploits a type of redundancy.

5. The method of claim 4 , wherein the type of redundancy comprises spatial redundancy.

6. The method of claim 4 , wherein the type of prediction comprises intra prediction.

7. The method of claim 4 , wherein the type of redundancy comprises temporal redundancy.

8. The method of claim 4 , wherein the type of prediction comprises inter prediction.

9. The method of claim 3 , wherein the partitions of the original probability lookup table are partitioned based on one or more transform unit sizes defined by the video coding standard.

10. The method of claim 9 , wherein the one or more transform unit sizes comprise 4×4 and 8×8.

11. A system, comprising:

a plurality of different probability lookup tables implemented in hardware, wherein each of the plurality of different probability lookup tables specifically corresponds to a different prediction mode of a video codec; and

an application-specific integrated circuit compute unit configured to:

for each candidate prediction mode among the different prediction modes of the video codec:

select one of the plurality of different probability lookup tables that corresponds to the candidate prediction mode;

use the selected one of the plurality of different probability lookup tables to calculate a corresponding token rate for the candidate prediction mode for a video to be encoded; and

use the corresponding token rate for the candidate prediction mode to determine a rate distortion cost for the video; and

encode the video using a selected one of the different prediction modes determined based on the determined rate distortion costs.

12. The system of claim 11 , wherein a prediction mode comprises one or more parameters defined by a standardized video coding format, wherein the one or more parameters comprise a transform unit size, and wherein the standardized video coding format is selected from a group consisting of: H.262, H.264, H.265, Theora, RealVideo RV40, VP9, and AV1.

13. The system of claim 11 , wherein each of the plurality of different probability lookup tables comprises a partition of an original probability lookup table that is partitioned based on one or more parameters defined by a standardized video coding format, and wherein the standardized video coding format is selected from a group consisting of: H.262, H.264, H.265, Theora, RealVideo RV40, VP9, and AV1.

14. The system of claim 13 , wherein the partitions of the original probability lookup table are partitioned based on a type of prediction that exploits a type of redundancy.

15. The system of claim 14 , wherein the type of redundancy comprises spatial redundancy.

16. The system of claim 14 , wherein the type of prediction comprises intra prediction.

17. The system of claim 14 , wherein the type of redundancy comprises temporal redundancy.

18. The system of claim 14 , wherein the type of prediction comprises inter prediction.

19. The system of claim 13 , wherein the partitions of the original probability lookup table are partitioned based on one or more transform unit sizes defined by the video coding standard.

20. A system, comprising:

A processor comprising logic to:

for each candidate prediction mode among different prediction modes of a video codec:

select one of a plurality of different probability lookup tables that corresponds to the candidate prediction mode;

use the selected one of the plurality of different probability lookup tables to calculate a corresponding token rate for the candidate prediction mode for a video to be encoded; and

use the corresponding token rate for the candidate prediction mode to determine a rate distortion cost for the video; and

encode the video using a selected one of the different prediction modes determined based on the determined rate distortion costs; and

a memory coupled to the processor and configured to provide the processor with instructions.