Mode complexity based coding strategy selection
View Patent ↗A system may receive an input stream for a coding operation. The system may determine available coding modes for the coding operation. The system may include coding selection logic that may determine a coding mode in response to the based on the available selection of coding modes. The coding selection logic may use the selected coding mode to determine a coding strategy. The selection logic may send an indication of the selected coding mode and coding strategy to coding logic to support execution of the coding operation, which may use the selected coding mode and coding strategy.
1. A method performed at a coding device, the method comprising:
determining, at a selection circuitry, a coding mode from among available coding modes for performing a coding task on a coding input, the coding mode indicating a block size for the coding task;
responsive to determining the coding mode, determining a coding strategy, wherein the coding strategy indicates a bit depth to be used for a cost metric calculation, the bit depth selected based at least in part on the block size;
sending an indication of the coding mode and the coding strategy to a coding circuitry;
initiating the cost metric calculation, at the coding circuitry, using the block size indicated by the coding mode and using the bit depth indicated by the coding strategy; and
performing, by the coding circuitry, the coding task using the coding mode and the coding strategy based at least in part on the cost metric calculation.
2. The method of claim 1 , wherein determining the coding mode comprises determining the block size for the coding task.
3. The method of claim 1 , wherein determining the coding mode comprises determining a codec for the coding task.
4. The method of claim 1 , wherein initiating the cost metric calculation comprises determining a rate-distortion optimization cost function.
5. The method of claim 1 , wherein determining the coding strategy comprises selecting the bit depth based at least in part on the block size.
6. The method of claim 5 , wherein selecting the bit depth comprises selecting a higher bit depth for the coding mode than for a second coding mode, the second coding mode having a greater complexity than the coding mode.
7. The method of claim 1 , wherein the coding strategy is further determined responsive to available resources for coding.
8. The method of claim 1 , wherein the coding strategy comprises a selection of a calculation domain for a transformation used in the coding task.
9. The method of claim 8 , wherein the calculation domain comprises a frequency domain or a spatial domain.
10. The method of claim 1 , wherein the coding task comprises decoding, encoding, transcoding, or any combination thereof.
11. A coding device comprising:
an input interface configured to receive an input stream;
a coding circuitry in data communication with the input interface and configured to perform a cost metric calculation on a portion of the input stream and to perform a coding task on the input stream; and
a selection circuitry, in data communication with the coding circuitry, the selection circuitry configured to:
determine a coding mode for the input stream, the coding mode indicating a block size for a block used in the coding task;
responsive to the determined coding mode, select a bit depth for the block based at least in part on the block size indicated by the coding mode; and
send an indication of the bit depth and the coding mode to coding circuitry to perform the cost metric calculation based at least in part on the block size and the bit depth in support of execution of the coding task based on the coding mode.
12. The coding device of claim 11 , wherein the selection circuitry is configured to determine the block size indicated by the coding mode.
13. The coding device of claim 11 , wherein the selection circuitry is configured to determine a codec for the coding task to determine the coding mode.
14. The coding device of claim 11 , wherein the selection circuitry is configured to determine the bit depth for the cost metric calculation.
15. The coding device of claim 11 , wherein the coding circuitry is configured to determine a rate-distortion optimization cost function to perform the cost metric calculation.
16. The coding device of claim 11 , wherein the selection circuitry is configured to select the bit depth based at least in part on an available resource for coding.
17. A method performed by a coding device, the method comprising:
receiving an input stream at a data interface on which a coding task is to be performed;
at selection circuitry in data communication with the data interface, determining a block size for a cost metric calculation to be performed on the input stream;
responsive to the determined block size, selecting a calculation domain for the cost metric calculation based at least in part on the determined block size; and
sending an indication of the calculation domain selection and the block size to coding circuitry to perform the cost metric calculation in the calculation domain based at least in part on the block size in support of executing the coding task.
18. The method of claim 17 , wherein the selected calculation domain comprises a frequency domain or a spatial domain.
19. The method of claim 18 , wherein selecting the calculation domain for the cost metric calculation based at least in part on the determined block size further comprises:
when the block size is above a threshold, selecting the frequency domain; and
when the block size is below the threshold, selecting the spatial domain.
20. The method of claim 17 , wherein the cost metric calculation comprises a determination of a rate-distortion optimization cost function to be performed within the selected calculation domain.