Abstract
Monte Carlo rendering algorithms often utilize correlations between pixels to improve efficiency and enhance image quality. For real-time applications in particular, repeated reservoir resampling offers a powerful framework to reuse samples both spatially in an image and temporally across multiple frames. While such techniques achieve equal-error up to 100× faster for real-time direct lighting [Bitterli et al. 2020] and global illumination [Ouyang et al. 2021; Lin et al. 2021], they are still far from optimal. For instance, spatiotemporal resampling often introduces noticeable correlation artifacts, while reservoirs holding more than one sample suffer from impoverishment in the form of duplicate samples. We demonstrate how interleaving Markov Chain Monte Carlo (MCMC) mutations with reservoir resampling helps alleviate these issues, especially in scenes with glossy materials and difficult-to-sample lighting. Moreover, our approach does not introduce any bias, and in practice, we find considerable improvement in image quality with just a single mutation per reservoir sample in each frame. u
ReSTIR PT [Lin et al. 2022] vs. ReSTIR PT with mutation (ours)
A single sample per pixel (spp) comparison of indirect illumination rendered using ReSTIR Path Tracing (PT) [Lin et al. 2022] with and without our sample mutations. By performing even a single mutation per sample, our approach can suppress correlation artifacts that may arise within ReSTIR samplers due to spatiotemporal reuse. Mutations improve visual fidelity of both rendered and denoised results (with the OptiX denoiser [NVIDIA 2017]) while leaving mean squared error unchanged.
Video
BibTeX
@article{sawhney2024,
author = {Sawhney, Rohan and Lin, Daqi and Kettunen, Markus and Bitterli, Benedikt and Ramamoorthi, Ravi and Wyman, Chris and Pharr, Matt},
title = {Decorrelating ReSTIR Samplers via MCMC Mutations},
year = {2024},
issue_date = {February 2024},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
volume = {43},
number = {1},
issn = {0730-0301},
url = {https://doi.org/10.1145/3629166},
doi = {10.1145/3629166},
journal = {ACM Trans. Graph.},
month = {jan},
articleno = {10},
numpages = {15},
keywords = {Markov Chain Monte Carlo, weighted reservoir sampling, resampled importance sampling, Real-time rendering}
}