Fine grained interleaved rendering applications in path tracing for cloud computing environments
Systems and methods of the present disclosure relate to fine grained interleaved rendering applications in path tracing for cloud computing environments. For example, a renderer and a rendering process may be employed for ray or path tracing and image-space filtering that interleaves the pixels of a frame into partial image fields and corresponding reduced-resolution images that are individually processed in parallel. Parallelization techniques described herein may allow for high quality rendered frames in less time, thereby reducing latency (or lag, in gaming applications) in high performance applications.
1. A system comprising:
one or more cloud-based servers including one or more graphics processing unit (GPU) resources, the one or more GPU resources comprising processing circuitry to:
generate, using one or more reflection rays corresponding to an application of one or more light transport simulation algorithms, and based at least on a first image field corresponding to a first subset of interleaved pixels of a full frame from a virtual scene, a first reduced resolution image corresponding to the first subset of interleaved pixels of the full frame;
generate, using the one or more refraction rays corresponding to the application of one or more light transport simulation algorithms, and based at least on a second image field corresponding to a second subset of interleaved pixels from the full frame, a second reduced resolution image corresponding to the second subset of interleaved pixels of the full frame; and
reconstruct the full frame based at least on the first reduced resolution image and the second reduced resolution image.
2. The system of claim 1 , wherein the generation of the first reduced resolution image includes using a first GPU resource of the one or more GPU resources and the generation of the second reduced resolution image includes using a second GPU resource of the one or more GPU resources.
3. The system of claim 1 , wherein
the generation of the first reduced resolution image is executed in parallel with the generation of the second reduced resolution image.
4. The system of claim 1 , wherein the one or more light transport simulation algorithms include at least one of a ray-tracing algorithm or a path-tracing algorithm.
5. The system of claim 1 , wherein the full frame is further reconstructed using one or more image-space filters, the one or more image-space filters including at least one of a denoising operation, a spatial filter, or a neighborhood filter.
6. The system of claim 1 , wherein the first subset of interleaved pixels includes every other pixel in at least three rows of pixels of the full frame starting with a first pixel, and the second subset of interleaved pixels includes every other pixel in the at least three rows of pixels of the full frame starting with a second pixel after the first pixel.
7. The system of claim 1 , wherein the first subset of interleaved pixels includes every other pixel in at least three columns of pixels of the full frame starting with a first pixel, and the second subset of interleaved pixels includes every other pixel in the at least three columns of pixels of the full frame starting with a second pixel after the first pixel.
8. The system of claim 1 , wherein the processing circuitry is further to encode the full frame to generate an encoded frame, and transmit the encoded frame to one or more end-user devices.
9. The system of claim 1 , wherein the system is comprised in at least one of:
a cloud-based system;
a system for cloud video game streaming;
a system for hosting one or more cloud virtual reality (VR) applications;
a system for executing one or more simulation applications; or
a system for hosting one or more remote desktop applications.
10. A processor comprising:
processing circuitry to:
generate, using one or more reflection rays corresponding to an application of one or more light transport simulation algorithms, and based at least on a first image field corresponding to a first subset of interleaved pixels of a full frame from a virtual scene, a first reduced resolution image corresponding to the first subset of interleaved pixels of the full frame;
generate, using one or more refraction rays corresponding to the application of one or more light transport simulation algorithms, and based at least on a second image field corresponding to a second subset of interleaved pixels from the full frame, a second reduced resolution image corresponding to the second subset of interleaved pixels of the full frame; and
generate, based at least on the first reduced resolution image and the second reduced resolution image, the full frame.
11. The processor of claim 10 , wherein the generation of the first reduced resolution image includes using a first GPU resource of the processor and the generation of the second reduced resolution image includes using a second GPU resource of the processor.
12. The processor of claim 10 , wherein at least one of:
the generation of the first reduced resolution image is executed in parallel with the generation of the second reduced resolution image.
13. The processor of claim 10 , wherein the one or more light transport simulation algorithms include at least one of a ray-tracing algorithm or a path-tracing algorithm.
14. The processor of claim 10 , wherein the generation of the full frame further uses one or more image-space filters, the one or more image-space filters including at least one of a denoising operation, a spatial filter, or a neighborhood filter.
15. The processor of claim 10 , wherein at least one of:
the first subset of interleaved pixels includes every other pixel in at least three rows of pixels of the full frame starting with a first pixel, and the second subset of interleaved pixels includes every other pixel in the at least throw rows of pixels of the full frame starting with a second pixel after the first pixel; or
the first subset of interleaved pixels includes every other pixel in at least three columns of pixels of the full frame starting with a first pixel, and the second subset of interleaved pixels includes every other pixel in the at least throw columns of pixels of the full frame starting with a second pixel after the first pixel.
16. The processor of claim 10 , wherein the processing circuitry is further to encode the full frame to generate an encoded frame, and transmit the encoded frame to one or more end-user devices.
17. The processor of claim 10 , wherein the processor is comprised in a cloud-based server.
18. The processor of claim 10 , wherein the processor is comprised in at least one of:
a cloud-based system;
a system for cloud video game streaming;
a system for hosting one or more cloud virtual reality (VR) applications;
a system for executing one or more simulation applications; or
a system for hosting one or more remote desktop applications.
19. A method comprising:
generating, using one or more reflection rays corresponding to an application of one or more light transport simulation algorithms, and based at least on a first image field corresponding to a first subset of interleaved pixels of a full frame from a virtual scene, a first reduced resolution image corresponding to the first subset of interleaved pixels of the full frame;
generating, using one or more refraction rays corresponding to the application of one or more light transport simulation algorithms, and based at least on a second image field corresponding to a second subset of interleaved pixels from the full frame, a second reduced resolution image corresponding to the second subset of interleaved pixels of the full frame; and
generating the full frame based at least on the first reduced resolution image and the second reduced resolution image.
20. The method of claim 19 , further comprising generating, based at least on the full frame, the first image field corresponding to the first subset of interleaved pixels and the second image field corresponding to the second subset of interleaved pixels.