Enhanced guiding for converging denoisers
The present embodiments generally relate to enhancing a quality of denoising in a real-time pathtracer. A pathtracer can implement a Monte-Carlo process, which can require many samples to converge to an error-acceptable result. Denoisers are often used in real-time (or near real-time), in order to accelerate the convergence process. The present embodiments further comprise systems and methods relating to an adaptive guidance algorithm to be implemented on top of various denoising algorithms (e.g., SVGF/ASVGF). The systems and methods as described herein can improve the gradient estimation quality, producing less blurry images. The systems and methods can also substantially assist many denoising algorithms in complicated illumination scenarios, which are common in CADs, such as indirect illuminated normal maps, specular paths, sharp specular reflections/refractions, very noisy direct illumination like dome lights, specular-specular sub-paths, sub surface scattering or hair, etc.
1 . A computer-implemented method comprising:
obtaining an initial image rendering to be processed to render a final image, wherein the initial image rendering includes information relating to unfiltered illuminations and features;
generating a guidance map for the initial image rendering based on the features, the guidance map controlling a level of moderation of variance-based filtering;
performing a variance-based filtering to the unfiltered illuminations using the guidance map to generate filtered illuminations; and
combining the filtered illuminations to generate the final image.
2 . The computer-implemented method of claim 1 , wherein the features comprise any of: surface and microsurface normals, a roughness, an object identifier, a material type, and a position derivative.
3 . The computer-implemented method of claim 1 , wherein the variance-based filtering comprises adaptive spatiotemporal variance-guided filtering (A-SVGF).
4 . The computer-implemented method of claim 1 , wherein the unfiltered illuminations and/or the filtered illuminations comprise any of: a direct illumination, an indirect illumination, and an indirect specular illumination.
5 . The computer-implemented method of claim 1 , wherein the final image filtered using the guidance map comprises a lower blurriness relating to any final image filtered without using the guidance map.
6 . The computer-implemented method of claim 1 , wherein the guidance map is configured to support a high-quality indirect illumination of one or more microsurfaces, as the guidance map is configured to signal that a high variance of any of the one or more microsurfaces does not need to be overblurred.
7 . The computer-implemented method of claim 1 , wherein the guidance map is configured to prevent overblurring of dome-lights and other low quality direct illumination sources.
8 . The computer-implemented method of claim 1 , wherein the guidance map is configured to be disabled and enabled.
9 . The computer-implemented method of claim 1 , wherein the guidance map is configured to speed up a convergence rate of sub surface scattered materials.
10 . A system comprising:
one or more processors;
one or more non-transitory processor readable storage devices comprising instructions which, when executed by the one or more processors, cause the one or more processor to perform operations comprising:
obtaining an initial image rendering that includes information relating to unfiltered illuminations and features, wherein the features comprise any of: surface and microsurface normals, a roughness, an object identifier, a material type, and a position derivative;
generating a history quality metric based on the features;
generating a guidance map for the initial image rendering based on the features and on the history quality metric, the guidance map controlling a level of moderation of variance-based filtering;
performing a variance-based filtering to the unfiltered illuminations using the guidance map to generate filtered illuminations; and
combining the filtered illuminations to generate a final image.
11 . The system of claim 10 , wherein the variance-based filtering comprises adaptive spatiotemporal variance-guided filtering (A-SVGF).
12 . The system of claim 10 , wherein the unfiltered illuminations and/or the filtered illuminations comprise any of: a direct illumination, an indirect illumination, and an indirect specular illumination.
13 . The system of claim 10 , wherein the final image filtered using the guidance map comprises a lower blurriness relating to any final image filtered without using the guidance map.
14 . The system of claim 10 , wherein the guidance map is configured to support a high-quality indirect illumination of one or more microsurfaces, as the guidance map is configured to signal that a high variance of any of the one or more microsurfaces does not need to be overblurred.
15 . The system of claim 10 , wherein the guidance map is configured to prevent overblurring of dome-lights and other low quality direct illumination sources.
16 . The system of claim 10 , wherein the guidance map is configured to be disabled and enabled.
17 . The system of claim 10 , wherein the guidance map is configured to speed up a convergence rate of sub surface scattered materials.
18 . One or more non-transitory computer-readable media comprising instructions which, when executed by one or more processors, cause the one or more processors to perform operations including:
obtaining an initial image rendering to be processed to render a final image, wherein the initial image rendering includes information relating to unfiltered illuminations and features;
generating a history quality metric based on the features;
generating a guidance map for the initial image rendering based on the features and the history quality metric, the guidance map controlling a level of moderation of variance-based filtering;
performing a variance-based filtering to the unfiltered illuminations using the guidance map to generate filtered illuminations; and
combining the filtered illuminations to generate the final image.
19 . The one or more non-transitory computer-readable media of claim 18 , wherein the unfiltered illuminations and/or the filtered illuminations comprise any of: a direct illumination, an indirect illumination, and an indirect specular illumination.
20 . The one or more non-transitory computer-readable media of claim 18 , wherein the final image filtered using the guidance map comprises a lower blurriness relating to any final image filtered without using the guidance map.