“Conditional Resampled Importance Sampling and ReSTIR” by Kettunen, Lin, Ramamoorthi and Wyman
Conference:
Type(s):
Title:
- Conditional Resampled Importance Sampling and ReSTIR
Session/Category Title: What're Your Points?
Presenter(s)/Author(s):
Abstract:
Recent work on generalized resampled importance sampling (GRIS) enables importance-sampled Monte Carlo integration with random variable weights replacing the usual division by probability density. This enables very flexible spatiotemporal sample reuse, even if neighboring samples (e.g., light paths) have intractable probability densities. Unlike typical Monte Carlo integration, which samples according to some PDF, GRIS instead resamples existing samples. But resampling with GRIS assumes samples have tractable marginal contribution weights, which is problematic if reusing, for example, light subpaths from unidirectionally-sampled paths. Reusing such subpaths requires conditioning by (non-reused) segments of the path prefixes. In this paper, we extend GRIS to conditional probability spaces, showing correctness given certain conditional independence between integration variables and their unbiased contribution weights. We show proper conditioning when using GRIS over randomized conditional domains and how to formulate a joint unbiased contribution weight for unbiased integration. To show our theory has practical impact, we prototype a modified ReSTIR PT with a final gather pass. This reuses subpaths, postponing reuse at least one bounce along each light path. As in photon mapping, such a final gather reduces blotchy artifacts from sample correlation and reduced correlation improves the behavior of modern denoisers on ReSTIR PT signals.
References:
[1]
Pablo Bauszat, Victor Petitjean, and Elmar Eisemann. 2017. Gradient-Domain Path Reusing. ACM Transactions on Graphics 36, 6 (2017), 229:1–229:9. https://doi.org/10.1145/3130800.3130886
[2]
Philippe Bekaert, Mateu Sbert, and John H Halton. 2002. Accelerating Path Tracing by Re-Using Paths. In Eurographics Workshop on Rendering. 125–134. https://doi.org/10.2312/EGWR.EGWR02.125-134
[3]
Nikolaus Binder, Sascha Fricke, and Alexander Keller. 2019. Massively parallel path space filtering. (2019). arxiv:1902.05942 [cs.GR]
[4]
Benedikt Bitterli, Chris Wyman, Matt Pharr, Peter Shirley, Aaron Lefohn, and Wojciech Jarosz. 2020. Spatiotemporal Reservoir Resampling for Real-time Ray Tracing with Dynamic Direct Lighting. ACM Transactions on Graphics 39, 4 (2020), 148:1–148:17. https://doi.org/10.1145/3386569.3392481
[5]
Min-Te Chao. 1982. A general purpose unequal probability sampling plan. Biometrika 69, 3 (1982), 653–656. https://doi.org/10.2307/2336002
[6]
Xi Deng, Miloš Hašan, Nathan Carr, Zexiang Xu, and Steve Marschner. 2021. Path Graphs: Iterative Path Space Filtering. ACM Transactions on Graphics 40, 6 (2021), 276:1–276:15. https://doi.org/10.1145/3478513.3480547
[7]
Xi Deng, Shaojie Jiao, Benedikt Bitterli, and Wojciech Jarosz. 2019. Photon Surfaces for Robust, Unbiased Volumetric Density Estimation. ACM Transactions on Graphics 38, 4 (2019), 46:1–46:12. https://doi.org/10.1145/3306346.3323041
[8]
Luc Devroye. 1986. Non-Uniform Random Variate Generation. Springer-Verlag. https://doi.org/10.1007/978-1-4613-8643-8
[9]
Víctor Elvira, Luca Martino, David Luengo, and Mónica F. Bugallo. 2019. Generalized Multiple Importance Sampling. Statistical Science 34, 1 (2019), 129–155. https://doi.org/10.1214/18-STS668
[10]
Iordanis Evangelou, Georgios Papaioannou, Konstantinos Vardis, and Andreas A. Vasilakis. 2021. Fast radius search exploiting ray-tracing frameworks. Journal of Computer Graphics Techniques 10, 1 (2021). https://jcgt.org/published/0010/01/02/
[11]
Binh-Son Hua, Adrien Gruson, Victor Petitjean, Matthias Zwicker, Derek Nowrouzezahrai, Elmar Eisemann, and Toshiya Hachisuka. 2019. A Survey on Gradient-Domain Rendering. 38, 2 (2019), 455–472. https://doi.org/10.1111/cgf.13652
[12]
Henrik Wann Jensen. 2001. Realistic Image Synthesis Using Photon Mapping. A. K. Peters, Ltd., USA.
[13]
Simon Kallweit, Petrik Clarberg, Craig Kolb, Tomáš Davidovič, Kai-Hwa Yao, Theresa Foley, Yong He, Lifan Wu, Lucy Chen, Tomas Akenine-Möller, Chris Wyman, Cyril Crassin, and Nir Benty. 2022. The Falcor Rendering Framework. https://github.com/NVIDIAGameWorks/Falcor [Online; accessed 22-August-2023].
[14]
Alexander Keller, Ken Dahm, and Nikolaus Binder. 2014. Path Space Filtering. In ACM SIGGRAPH Talks. Article 68. https://doi.org/10.1145/2614106.2614149
[15]
Eric P. Lafortune and Yves D. Willems. 1993. Bi-Directional Path Tracing. In Proceedings of Compugraphics. 145–153. https://graphics.cs.kuleuven.be/publications/BDPT/
[16]
Daqi Lin, Markus Kettunen, Benedikt Bitterli, Jacopo Pantaleoni, Cem Yuksel, and Chris Wyman. 2022. Generalized Resampled Importance Sampling: Foundations of ReSTIR. ACM Transactions on Graphics 41, 4 (2022), 75:1–75:23. https://doi.org/10.1145/3528223.3530158
[17]
Daqi Lin, Chris Wyman, and Cem Yuksel. 2021. Fast Volume Rendering with Spatiotemporal Reservoir Resampling. ACM Transactions on Graphics 40, 6 (2021), 279:1–279:16. https://doi.org/10.1145/3478513.3480499
[18]
Zander Majercik, Thomas Müller, Alexander Keller, Derek Nowrouzezahrai, and Morgan McGuire. 2021. Dynamic Diffuse Global Illumination Resampling. In ACM SIGGRAPH Talks. Article 24. https://doi.org/10.1145/3450623.3464635
[19]
NVIDIA. 2021. RTX Direct Illumination API. https://developer.nvidia.com/rtxdi. [Online; accessed 22-August-2023].
[20]
NVIDIA. 2023. NVIDIA DLSS 3.5: Enhancing Ray Tracing With AI. https://www.nvidia.com/en-us/geforce/news/nvidia-dlss-3-5-ray-reconstruction. [Online; accessed 12-September-2023].
[21]
Yaobin Ouyang, Shiqiu Liu, Markus Kettunen, Matt Pharr, and Jacopo Pantaleoni. 2021. ReSTIR GI: Path Resampling for Real-Time Path Tracing. Computer Graphics Forum 40, 8 (2021), 17–29. https://doi.org/10.1111/cgf.14378
[22]
Orion Pobursky. 2021. Tower Bridge Model (CC BY-2.0) available on LDraw.org. http://omr.ldraw.org/files/1692 [Online; accessed 22-August-2023].
[23]
Rohan Sawhney, Daqi Lin, Markus Kettunen, Benedikt Bitterli, Ravi Ramamoorthi, Chris Wyman, and Matt Pharr. 2022. Decorrelating ReSTIR Samplers via MCMC Mutations. arxiv:2211.00166 [cs.GR]
[24]
Benjamin Segovia, Jean Claude Iehl, Richard Mitanchey, and Bernard Péroche. 2006. Bidirectional Instant Radiosity. In Eurographics Symposium on Rendering. 389–397. https://doi.org/10.2312/EGWR/EGSR06/389-397
[25]
Justin Talbot, David Cline, and Parris Egbert. 2005. Importance Resampling for Global Illumination. In Eurographics Symposium on Rendering. 139–146. https://doi.org/10.2312/EGWR/EGSR05/139-146
[26]
Lorenzo Tessari, Johannes Hanika, and Carsten Dachsbacher. 2017. Local quasi-monte carlo exploration. In Eurographics Symposium on Rendering: Experimental Ideas & Implementations. 71–81. https://doi.org/10.2312/sre.20171196
[27]
Eric Veach and Leonidas Guibas. 1995a. Bidirectional estimators for light transport. In Photorealistic Rendering Techniques. Springer, 145–167. https://doi.org/10.1007/978-3-642-87825-1_11
[28]
Eric Veach and Leonidas J. Guibas. 1995b. Optimally Combining Sampling Techniques for Monte Carlo Rendering. In Proceedings of the 22nd Annual Conference on Computer Graphics and Interactive Techniques. 419–428. https://doi.org/10.1145/218380.218498
[29]
Eric Veach and Leonidas J. Guibas. 1997. Metropolis Light Transport. In Proceedings of the 24th Annual Conference on Computer Graphics and Interactive Techniques. 65–76. https://doi.org/10.1145/258734.258775
[30]
Andrea Weidlich, Chloe LeGendre, Carlos Aliaga, Christophe Hery, Jean-Marie Aubry, Jiří Vorba, Daniele Siragusano, and Richard Kirk. 2022. Practical Aspects of Spectral Data in Digital Content Production. In ACM SIGGRAPH Courses. Article 11. https://doi.org/10.1145/3532720.3535632
[31]
Rex West, Iliyan Georgiev, Adrien Gruson, and Toshiya Hachisuka. 2020. Continuous multiple importance sampling. ACM Transactions on Graphics 39, 4 (2020), 136:1–136:12. https://doi.org/10.1145/3386569.3392436
[32]
Rex West, Iliyan Georgiev, and Toshiya Hachisuka. 2022. Marginal Multiple Importance Sampling. In ACM SIGGRAPH Asia Conference Papers. 42:1–42:8. https://doi.org/10.1145/3550469.3555388
[33]
Chris Wyman and Alexey Panteleev. 2021. Rearchitecting Spatiotemporal Resampling for Production. In ACM/EG Symposium on High Perfrormance Graphics. 23–41. https://doi.org/10.2312/hpg.20211281
