Boundary Value Caching for Walk on Spheres

Grid-free Monte Carlo methods such as walk on spheres can be used to solve elliptic partial differential equations without mesh generation or global solves. However, such methods independently estimate the solution at every point, and hence do not take advantage of the high spatial regularity of solutions to elliptic problems. We propose a fast caching strategy which first estimates solution values and derivatives at randomly sampled points along the boundary of the domain (or a local region of interest). These cached values then provide cheap, output-sensitive evaluation of the solution (or its gradient) at interior points, via a boundary integral formulation. Unlike classic boundary integral methods, our caching scheme introduces zero statistical bias and does not require a dense global solve. Moreover we can handle imperfect geometry (e.g., with self-intersections) and detailed boundary/source terms without repairing or resampling the boundary representation. Overall, our scheme is similar in spirit to virtual point light methods from photorealistic rendering: it suppresses the typical salt-and-pepper noise characteristic of independent Monte Carlo estimates, while still retaining the many advantages of Monte Carlo solvers: progressive evaluation, trivial parallelization, geometric robustness, etc. We validate our approach using test problems from visual and geometric computing.

[1]  Keenan Crane,et al.  Walk on Stars: A Grid-Free Monte Carlo Method for PDEs with Neumann Boundary Conditions , 2023, ACM Trans. Graph..

[2]  Derek Nowrouzezahrai,et al.  A Monte Carlo Method for Fluid Simulation , 2022, ACM Trans. Graph..

[3]  Wenzel Jakob,et al.  Solving Inverse PDE Problems using Grid-Free Monte Carlo Estimators , 2022, ArXiv.

[4]  Wojciech Jarosz,et al.  A bidirectional formulation for Walk on Spheres , 2022, Comput. Graph. Forum.

[5]  Keenan Crane,et al.  Grid-free Monte Carlo for PDEs with spatially varying coefficients , 2022, ACM Trans. Graph..

[6]  M. Pharr,et al.  ReSTIR GI: Path Resampling for Real‐Time Path Tracing , 2021, Comput. Graph. Forum.

[7]  Aaron E. Lefohn,et al.  Spatiotemporal reservoir resampling for real-time ray tracing with dynamic direct lighting , 2020, ACM Trans. Graph..

[8]  Cem Yuksel,et al.  Real-Time Stochastic Lightcuts , 2020, Proc. ACM Comput. Graph. Interact. Tech..

[9]  Jerry Jinfeng Guo,et al.  Geometric Sample Reweighting for Monte Carlo Integration , 2019, Comput. Graph. Forum.

[10]  Cem Yuksel,et al.  Stochastic lightcuts , 2019, High Performance Graphics.

[11]  Alec Jacobson,et al.  Fast winding numbers for soups and clouds , 2018, ACM Trans. Graph..

[12]  W. Hackbusch,et al.  Hierarchical Matrices: Algorithms and Analysis , 2015 .

[13]  Adam Arbree,et al.  Scalable Realistic Rendering with Many‐Light Methods , 2014, Eurographics.

[14]  H. Jensen,et al.  Stochastic progressive photon mapping , 2009, ACM Trans. Graph..

[15]  Shinji Ogaki,et al.  Progressive photon mapping , 2008, ACM Trans. Graph..

[16]  Mark Braverman,et al.  The rate of convergence of the Walk on Spheres Algorithm , 2008, 0810.3343.

[17]  Bernard Péroche,et al.  Bidirectional instant radiosity , 2006, EGSR '06.

[18]  Adam Arbree,et al.  To appear in the ACM SIGGRAPH conference proceedings Lightcuts: A Scalable Approach to Illumination , 2022 .

[19]  Greg Humphreys,et al.  Physically Based Rendering: From Theory to Implementation , 2004 .

[20]  Wen Chen,et al.  Symmetric boundary knot method , 2002, ArXiv.

[21]  Graeme Fairweather,et al.  The method of fundamental solutions for elliptic boundary value problems , 1998, Adv. Comput. Math..

[22]  Alexander Keller,et al.  Instant radiosity , 1997, SIGGRAPH.

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

[24]  Paul Gibbon,et al.  Many-body tree methods in physics , 1996 .

[25]  Leslie Greengard,et al.  A fast algorithm for particle simulations , 1987 .

[26]  Martin Costabel,et al.  Principles of boundary element methods , 1987 .

[27]  Alastair J. Walker,et al.  An Efficient Method for Generating Discrete Random Variables with General Distributions , 1977, TOMS.

[28]  A. J. Walker New fast method for generating discrete random numbers with arbitrary frequency distributions , 1974 .

[29]  M. E. Muller Some Continuous Monte Carlo Methods for the Dirichlet Problem , 1956 .

[30]  Keenan Crane,et al.  Monte Carlo geometry processing , 2020, ACM Trans. Graph..

[31]  T. Kollig,et al.  Illumination in the Presence of Weak Singularities , 2006 .

[32]  P. Hunter,et al.  FEM/BEM NOTES , 2001 .

[33]  Yinglong Zhang,et al.  On the dual reciprocity boundary element method , 1993 .

[34]  O. Widlund Domain Decomposition Algorithms , 1993 .