Replica Exchange Light Transport

We solve the light transport problem by introducing a novel unbiased Monte Carlo algorithm called replica exchange light transport, inspired by the replica exchange Monte Carlo method in the fields of computational physics and statistical information processing. The replica exchange Monte Carlo method is a sampling technique whose operation resembles simulated annealing in optimization algorithms using a set of sampling distributions. We apply it to the solution of light transport integration by extending the probability density function of an integrand of the integration to a set of distributions. That set of distributions is composed of combinations of the path densities of different path generation types: uniform distributions in the integral domain, explicit and implicit paths in light (particle/photon) tracing, indirect paths in bidirectional path tracing, explicit and implicit paths in path tracing, and implicit caustics paths seen through specular surfaces including the delta function in path tracing. The replica‐exchange light transport algorithm generates a sequence of path samples from each distribution and samples the simultaneous distribution of those distributions as a stationary distribution by using the Markov chain Monte Carlo method. Then the algorithm combines the obtained path samples from each distribution using multiple importance sampling. We compare the images generated with our algorithm to those generated with bidirectional path tracing and Metropolis light transport based on the primary sample space. Our proposing algorithm has better convergence property than bidirectional path tracing and the Metropolis light transport, and it is easy to implement by extending the Metropolis light transport.

[1]  Yves D. Willems,et al.  Rendering Participating Media with Bidirectional Path Tracing , 1996, Rendering Techniques.

[2]  Leonidas J. Guibas,et al.  Bidirectional Estimators for Light Transport , 1995 .

[3]  Henrik Wann Jensen,et al.  Global Illumination using Photon Maps , 1996, Rendering Techniques.

[4]  Donald P. Greenberg,et al.  Global Illumination via Density Estimation , 1995, Rendering Techniques.

[5]  Yves D. Willems,et al.  Reducing the number of shadow rays in bidirectional path tracing , 1995 .

[6]  James Arvo,et al.  Particle transport and image synthesis , 1990, SIGGRAPH.

[7]  Leonidas J. Guibas,et al.  Optimally combining sampling techniques for Monte Carlo rendering , 1995, SIGGRAPH.

[8]  James T. Kajiya,et al.  The rendering equation , 1986, SIGGRAPH.

[9]  Yu-Chi Lai,et al.  Photorealistic Image Rendering with Population Monte Carlo Energy Redistribution , 2007, Rendering Techniques.

[10]  Peter Shirley,et al.  A variance analysis of the Metropolis Light Transport algorithm , 2001, Comput. Graph..

[11]  Leonidas J. Guibas,et al.  Metropolis light transport , 1997, SIGGRAPH.

[12]  Yukito Iba.  EXTENDED ENSEMBLE MONTE CARLO , 2001 .

[13]  B. Segovia,et al.  Coherent Metropolis Light Transport with Multiple-Try Mutations , 2007 .

[14]  Wolfgang Heidrich,et al.  Bidirectional importance sampling for direct illumination , 2005, EGSR '05.

[15]  Wolfgang Heidrich,et al.  Sequential Sampling for Dynamic Environment Map Illumination , 2022 .

[16]  Gregory J. Ward,et al.  A ray tracing solution for diffuse interreflection , 2008, SIGGRAPH '08.

[17]  Parris K. Egbert,et al.  Importance resampling for global illumination , 2005, EGSR '05.

[18]  Hans-Peter Seidel,et al.  Ray maps for global illumination , 2004, SIGGRAPH '04.

[19]  Mateu Sbert,et al.  Go with the Winners Strategy in Path Tracing , 2005, J. WSCG.

[20]  Yukito IBA,et al.  Population Monte Carlo algorithms , 2000, cond-mat/0008226.

[21]  Alexander Keller,et al.  Metropolis Light Transport for Participating Media , 2000, Rendering Techniques.

[22]  Csaba Kelemen,et al.  Simple and Robust Mutation Strategy for Metropolis Light Transport Algorithm , 2001 .

[23]  Justin Talbot,et al.  Energy redistribution path tracing , 2005, ACM Trans. Graph..

[24]  Yves D. Willems,et al.  Bi-directional path tracing , 1993 .

[25]  Werner Purgathofer,et al.  On The Start-Up Bias Problem Of Metropolis Sampling , 1999 .