“Bayesian online regression for adaptive direct illumination sampling” by Vévoda, Kondapaneni and Křivánek
Conference:
Type:
Entry Number: 125
Session Title:
- Learning for Rendering and Material Acquisition
Title:
- Bayesian online regression for adaptive direct illumination sampling
Moderator(s):
Presenter(s)/Author(s):
Abstract:
Direct illumination calculation is an important component of any physically-based Tenderer with a substantial impact on the overall performance. We present a novel adaptive solution for unbiased Monte Carlo direct illumination sampling, based on online learning of the light selection probability distributions. Our main contribution is a formulation of the learning process as Bayesian regression, based on a new, specifically designed statistical model of direct illumination. The net result is a set of regularization strategies to prevent over-fitting and ensure robustness even in early stages of calculation, when the observed information is sparse. The regression model captures spatial variation of illumination, which enables aggregating statistics over relatively large scene regions and, in turn, ensures a fast learning rate. We make the method scalable by adopting a light clustering strategy from the Lightcuts method, and further reduce variance through the use of control variates. As a main design feature, the resulting algorithm is virtually free of any preprocessing, which enables its use for interactive progressive rendering, while the online learning still enables super-linear convergence.
References:
1. Niels Billen, Björn Engelen, Ares Lagae, and Philip Dutre. 2013. Probabilistic Visibility Evaluation for Direct Illumination. Computer Graphics Forum 32, 4 (2013), 39–47. Google ScholarDigital Library
2. Christopher M. Bishop. 2006. Pattern Recognition and Machine Learning. Springer-Verlag New York Inc. Google ScholarDigital Library
3. Malik Boughida and Tamy Boubekeur. 2017. Bayesian collaborative denoising for Monte Carlo rendering. Computer Graphics Forum 36, 4 (2017), 137–153. Google ScholarDigital Library
4. Jonathan Brouillat, Christian Bouville, Brad Loos, Charles Hansen, and Kadi Bouatouch. 2009. A Bayesian Monte Carlo approach to global illumination. Computer Graphics Forum 28, 8 (2009), 2315–2329.Google ScholarCross Ref
5. Brian C. Budge, John C. Anderson, and Kenneth I. Joy. 2008. Caustic forecasting: Unbiased estimation of caustic lighting for global illumination. Computer Graphics Forum 27, 7 (2008), 1963–1970.Google ScholarCross Ref
6. Norbert Bus, Nabil H. Mustafa, and Venceslas Biri. 2015. IlluminationCut. Computer Graphics Forum (Proceedings of Eurographics 2015) 34, 2 (2015), 561–573. Google ScholarDigital Library
7. Olivier Cappé, Randal Douc, Arnaud Guillin, Jean-Michel Marin, and Christian P. Robert. 2008. Adaptive importance sampling in general mixture classes. Statistics and Computing 18, 4 (2008), 447–459. arXiv:0710.4242 Google ScholarDigital Library
8. Olivier Cappé, Arnaud Guillin, Jean-Michel Marin, and Christian P. Robert. 2004. Population Monte Carlo. Journal of Computational and Graphical Statistics 13, 4 (2004), 907–929.Google ScholarCross Ref
9. Petrik Clarberg and Tomas Akenine-Möller. 2008. Exploiting visibility correlation in direct illumination. Computer Graphics Forum 27, 4 (2008), 1125–1136. Google ScholarDigital Library
10. Jean-Marie Cornuet, Jean-Michel Marin, Antonietta Mira, and Christian P. Robert. 2009. Adaptive Multiple Importance Sampling. Scandinavian Journal of Statistics 39, 4 (2009), 798–812.Google ScholarCross Ref
11. Michael Donikian, Bruce Walter, Kavita Bala, Sebastian Fernandez, and Donald P. Greenberg. 2006. Accurate direct illumination using iterative adaptive sampling. IEEE Transactions on Visualization and Computer Graphics 12, 3 (2006), 353–363. Google ScholarDigital Library
12. Randal Douc and Arnaud Guillin. 2007. Minimum variance importance sampling via population Monte Carlo. ESAPM: Probability and Statistics 11 (2007), 427–447.Google ScholarCross Ref
13. Shaohua Fan, Yu-Chi Lai, Stephen Chenney, and Charles Dyer. 2007. Population Monte Carlo Sampler for Rendering. Technical Report 1613. Department of Computer Sciences, University of Wisconsin-Madison.Google Scholar
14. Sebastian Fernandez, Kavita Bala, and Donald P. Greenberg. 2002. Local Illumination Environments for Direct Lighting Acceleration. Proceedings of the 13th Eurographics workshop on Rendering (2002), 7–14. Google ScholarDigital Library
15. Manuel N. Gamito. 2016. Solid Angle Sampling of Disk and Cylinder Lights. Comput. Graph. Forum 35, 4 (July 2016), 25–36. Google ScholarDigital Library
16. Iliyan Georgiev, Jaroslav Křivánek, Stefan Popov, and Philipp Slusallek. 2012. Importance Caching for Complex Illumination. Computer Graphics Forum 31 (2012). Google ScholarDigital Library
17. David Hart, Philip Dutré, and Donald P. Greenberg. 1999. Direct illumination with lazy visibility evaluation. SIGGRAPH ’99 (1999), 147–154. Google ScholarDigital Library
18. Heinrich Hey and Werner Purgathofer. 2002. Importance sampling with hemispherical particle footprints. Proceedings of the 18th spring conference on Computer graphics – SCCG ’02 (2002), 107. Google ScholarDigital Library
19. Henrik Wann Jensen. 1995. Importance driven path tracing using the photon map. Rendering Techniques 95 (1995), 326–335.Google ScholarCross Ref
20. Henrik Wann Jensen and Niels Jørgen Christensen. 1995. Efficiently rendering shadows using the photon map. Compugraphics ’95 (1995), 285–291.Google Scholar
21. Malvin H. Kalos and Paula A. Whitlock. 1986. Monte Carlo Methods. Wiley-Interscience. Google ScholarDigital Library
22. Eric P. Lafortune and Yves D. Willems. 1995. A 5D Tree to Reduce the Variance of Monte Carlo Ray Tracing. (1995), 11–20.Google Scholar
23. Yu-Chi Lai, Shao Hua Fan, Stephen Chenney, and Charcle Dyer. 2007. Photorealistic Image Rendering with Population Monte Carlo Energy Redistribution. In Proceedings of Eurographics Symposium on Rendering (EGSR’07). Google ScholarDigital Library
24. Marc Lebrun, Antoni Buades, and Jean-Michel Morel. 2013. A Nonlocal Bayesian Image Denoising Algorithm. SIAM Journal on Imaging Sciences 6, 3 (2013), 1665–1688.Google ScholarCross Ref
25. Ricardo Marques, Christian Bouville, Mickaël Ribardiere, Luís Paulo Santos, and Kadi Bouatouch. 2013. A spherical gaussian framework for Bayesian Monte Carlo rendering of glossy surfaces. IEEE Trans. Vis. Comput. Graph. 19, 10 (2013), 1619–1632.Google ScholarCross Ref
26. Luca Martino, Víctor Elvira, David Luengo, and Jukka Corander. 2015. An Adaptive Population Importance Sampler: Learning from Uncertainty. IEEE Transactions on Signal Processing 63, 16 (2015), 4422–4437.Google ScholarDigital Library
27. Thomas Müller, Markus Gross, and Jan Novák. 2017. Practical Path Guiding for Efficient Light-Transport Simulation. Eurographics Symposium on Rendering 36, 4 (2017).Google Scholar
28. Art Owen and Yi Zhou. 2000. Safe and Effective Importance Sampling. J. Amer. Statist. Assoc. 95, 449 (2000), 135–143.Google ScholarCross Ref
29. Jacopo Pantaleoni and Eric Heitz. 2017. Notes on optimal approximations for importance sampling. 2, 5 (2017). arXiv:1707.08358Google Scholar
30. Eric Paquette, Pierre Poulin, and George Drettakis. 1998. A light hierarchy for fast rendering of scenes with many lights. Computer Graphics Forum 17, 3 (1998), 63–74.Google ScholarCross Ref
31. Vincent Pegoraro, Carson Brownlee, Peter S. Shirley, and Steven G. Parker. 2008. Towards interactive global illumination effects via sequential Monte Carlo adaptation. IEEE/EG Symposium on Interactive Ray Tracing 2008 (2008), 107–114.Google ScholarCross Ref
32. Matt Pharr, Wenzel Jakob, and Greg Humphreys. 2016. Physically Based Rendering, From Theory to Implementation (3rd ed.). Morgan Kaufmann Publishers Inc. Google ScholarDigital Library
33. Stefan Popov, Iliyan Georgiev, Philipp Slusallek, and Carsten Dachsbacher. 2013. Adaptive quantization visibility caching. Computer Graphics Forum 32, 2 (2013), 399–408.Google ScholarCross Ref
34. Carl Edward Rasmussen and Zoubin Ghahramani. 2003. Bayesian Monte Carlo. Advances in Neural Information Processing Systems 15 1 (2003), 489–496.Google Scholar
35. Fabrice Rousselle, Wojciech Jarosz, and Jan Novák. 2016. Image-space Control Variates for Rendering. ACM Transactions on Graphics 35, 6 (2016). Google ScholarDigital Library
36. Peter Shirley, Changyaw Wang, and Kurt Zimmerman. 1996. Monte Carlo techniques for direct lighting calculations. ACM Transactions on Graphics 15, 1 (1996), 1–36. Google ScholarDigital Library
37. Eric Veach. 1997. Robust Monte Carlo Methods for Light Transport Simulation. Dissertation at the Department of Computer Science of Stanford University (1997). Google ScholarDigital Library
38. Jiří Vorba, Ondřej Karlík, Martin Šik, Tobias Ritschel, and Jaroslav Křivánek. 2014. On-line Learning of Parametric Mixture Models for Light Transport Simulation. ACM Trans. Graph. 33, 4 (2014), 101:1–101:11. Google ScholarDigital Library
39. Ingo Wald and Carsten Benthin. 2003. Interactive Global Illumination in Complex and Highly Occluded Environments. Eurograhics Symposium on Rendering (2003), 1–9. Google ScholarDigital Library
40. Bruce Walter, Adam Arbree, Kavita Bala, and Donald P. Greenberg. 2006. Multidimensional lightcuts. ACM Transactions on Graphics 25, 3 (2006), 1081. Google ScholarDigital Library
41. Bruce Walter, Sebastian Fernandez, Adam Arbree, Kavita Bala, Michael Donikian, and Donald P Greenberg. 2005. Lightcuts: a scalable approach to illumination. ACM Transactions on Graphics 24, 3 (2005), 1098–1107. Google ScholarDigital Library
42. Rui Wang and Oskar Akerlund. 2009. Bidirectional Importance Sampling for Unstructured Direct Illumination. Computer Graphics Forum 28, 2 (2009), 269–278.Google ScholarCross Ref
43. Gregory J. Ward. 1994. Adaptive Shadow Testing for Ray Tracing. Proceedings of the Second Eurographics Workshop on Rendering (1994), 11–20.Google ScholarCross Ref
44. Yu Ting Wu and Yung Yu Chuang. 2013. VisibilityCluster: Average directional visibility for many-light rendering. IEEE Trans. Vis. Comput. Graph. 19, 9 (2013), 1566–1578. Google ScholarDigital Library
