Methods and apparatus for machine learning rendering
The present disclosure relates to methods and apparatus for graphics processing. The apparatus can obtain at least one input image including a plurality of pixels. Additionally, the apparatus can determine shading information for each of the plurality of pixels in the at least one input image. The apparatus can also determine a shading map based on the determined shading information for each of the plurality of pixels in the at least one input image. In some aspects, the apparatus can generate at least one output image based on the at least one input image and the determined shading map. The apparatus can also enhance a quality of the at least one output image. In some aspects, the quality of the at least one output image can be enhanced based on machine learning. Further, the apparatus can generate the at least one input image including the plurality of pixels.
1. A method for graphics processing, comprising:
obtaining at least one input image including a plurality of pixels;
determining shading information for each of the plurality of pixels in the at least one input image based on machine learning; and
determining a shading map based on the determined shading information for each of the plurality of pixels in the at least one input image;
wherein the machine learning makes decisions based on predicted quality of at least one output image and computational power to render the at least one output image.
2. The method of claim 1 , further comprising: generating the at least one output image based on the at least one input image and the determined shading map.
3. The method of claim 2 , wherein generating the at least one output image based on the at least one input image and the determined shading map further comprises:
rendering the at least one output image at a renderer based on the at least one input image and the determined shading map.
4. The method of claim 2 , further comprising:
enhancing a quality of the at least one output image.
5. The method of claim 4 , wherein the quality of the at least one output image is enhanced based on machine learning.
6. The method of claim 5 , wherein the quality of the at least one output image is enhanced based on machine learning at a deep neural network (DNN) component.
7. The method of claim 1 , wherein the shading information for each of the plurality of pixels in the at least one input image is determined based on machine learning at a DNN component.
8. The method of claim 1 , further comprising:
generating the at least one input image including the plurality of pixels.
9. The method of claim 1 , wherein the at least one input image includes a low resolution.
10. The method of claim 1 , wherein the at least one input image is based on at least one previously rendered image.
11. The method of claim 1 , wherein the at least one input image is based on at least one depth map.
12. The method of claim 1 , wherein the determined shading information for each of the plurality of pixels includes at least one of depth information, normal information, texture information, a per-pixel texture identification, or light visibility information.
13. The method of claim 1 , wherein the determined shading information for each of the plurality of pixels includes a rendering quality for the pixel based on the at least one input image.
14. The method of claim 1 , wherein the determined shading map includes a shading rate for each of the plurality of pixels based on variable rate shading (VRS).
15. The method of claim 1 , wherein the shading information for each of the plurality of pixels is determined in a graphics processing pipeline of a graphics processing unit (GPU).
16. The method of claim 1 , wherein the shading information for each of the plurality of pixels is determined at a central processing unit (CPU).
17. An apparatus for graphics processing, comprising:
a memory; and
at least one processor coupled to the memory and configured to:
obtain at least one input image including a plurality of pixels;
determine shading information for each of the plurality of pixels in the at least one input image based on machine learning; and
determine a shading map based on the determined shading information for each of the plurality of pixels in the at least one input image;
wherein the machine learning makes decisions based on predicted quality of at least one output image and computational power to render the at least one output image.
18. The apparatus of claim 17 , wherein the at least one processor is further configured to: generate the at least one output image based on the at least one input image and the determined shading map.
19. The apparatus of claim 18 , wherein to generate the at least one output image based on the at least one input image and the determined shading map includes the at least one processor further configured to:
render the at least one output image at a renderer based on the at least one input image and the determined shading map.
20. The apparatus of claim 18 , wherein the at least one processor is further configured to:
enhance a quality of the at least one output image.
21. The apparatus of claim 20 , wherein the quality of the at least one output image is enhanced based on machine learning.
22. The apparatus of claim 21 , wherein the quality of the at least one output image is enhanced based on machine learning at a deep neural network (DNN) component.
23. The apparatus of claim 17 , wherein the shading information for each of the plurality of pixels in the at least one input image is determined based on machine learning at a DNN component.
24. The apparatus of claim 17 , wherein the at least one processor is further configured to:
generate the at least one input image including the plurality of pixels.
25. The apparatus of claim 17 , wherein the at least one input image is based on at least one depth map.
26. The apparatus of claim 17 , wherein the determined shading information for each of the plurality of pixels includes at least one of depth information, normal information, texture information, a per-pixel texture identification, or light visibility information.
27. An apparatus for graphics processing, comprising:
means for obtaining at least one input image including a plurality of pixels;
means for determining shading information for each of the plurality of pixels in the at least one input image based on machine learning; and
means for determining a shading map based on the determined shading information for each of the plurality of pixels in the at least one input image;
wherein the machine learning makes decisions based on predicted quality of at least one output image and computational power to render the at least one output image.
28. A computer-readable medium storing computer executable code for graphics processing, comprising code to:
obtain at least one input image including a plurality of pixels;
determine shading information for each of the plurality of pixels in the at least one input image based on machine learning; and
determine a shading map based on the determined shading information for each of the plurality of pixels in the at least one input image;
wherein the machine learning makes decisions based on predicted quality of at least one output image and computational power to render the at least one output image.
29. The method of claim 1 , wherein the shading information for each of the plurality of pixels in the at least one input image is determined based on machine learning to reduce power consumption or increase rendering performance.
30. The apparatus of claim 17 , wherein the shading information for each of the plurality of pixels in the at least one input image is determined based on machine learning to reduce power consumption or increase rendering performance.