RGL
Realistic Graphics Lab
EPFL Logo

Path Space MCMC Methods in Computer Graphics

Book chapter Monte Carlo and Quasi-Monte Carlo Methods (Springer Proceedings in Mathematics & Statistics)

Il­lus­tra­tion of the loc­al en­ergy bal­ance equa­tion which de­scribes light trans­port on sur­faces. In the above ex­ample, it is used to com­pute the pixel col­or of the sur­face loc­a­tion high­lighted in white.

Abstract

The ob­ject­ive of a ren­der­ing al­gorithm is to com­pute a pho­to­graph of a sim­u­lated real­ity, which en­tails find­ing all the paths along which light can flow from a set of light sources to the cam­era. The pur­pose of this art­icle is to present a high-level over­view of the un­der­ly­ing phys­ics and ana­lyze how this leads to a high-di­men­sion­al in­teg­ra­tion prob­lem that is typ­ic­ally handled us­ing Monte Carlo meth­ods. Fol­low­ing this, we sur­vey re­cent work on path space Markov Chain Monte Carlo (MCMC) meth­ods that com­pute the res­ult­ing in­teg­rals us­ing pro­pos­al dis­tri­bu­tions defined on sets of light paths.

Text citation

Wenzel Jakob and Steve Marschner. 2016. Path Space MCMC Methods in Computer Graphics. In Monte Carlo and Quasi-Monte Carlo Methods (Springer Proceedings in Mathematics & Statistics).

BibTeX
@incollection{Jakob2016Path,
    author = {Wenzel Jakob and Steve Marschner},
    title = {Path Space MCMC Methods in Computer Graphics},
    booktitle = {Monte Carlo and Quasi-Monte Carlo Methods: MCQMC, Leuven, Belgium, April 2014 (Springer Proceedings in Mathematics & Statistics)},
    editor = {Ronald Cools and Dirk Nuyens},
    publisher = {Springer International Publishing},
    year = {2016},
    month = jul,
    doi = {10.1007/978-3-319-33507-0}
}