Memory-compact Metropolis light transport on GPUs
- Authors: Frolov V.A.1,2, Galaktionov V.A.1
-
Affiliations:
- The Keldysh Institute of Applied Mathematics
- Moscow State University
- Issue: Vol 43, No 3 (2017)
- Pages: 196-203
- Section: Article
- URL: https://journals.rcsi.science/0361-7688/article/view/176509
- DOI: https://doi.org/10.1134/S0361768817030057
- ID: 176509
Cite item
Abstract
Solutions to the key problems of Metropolis light transport implementation on GPUs are proposed. A “burn-in” method relying on the ordinary Monte Carlo method, owing to which the “startup bias” is significantly reduced, is suggested. Memory optimizations methods (including multiple proposal Metropolis light transport) are proposed, and technical aspects of efficient Metropolis light transport implementation on GPUs are discussed.
About the authors
V. A. Frolov
The Keldysh Institute of Applied Mathematics; Moscow State University
Author for correspondence.
Email: vova@frolov.pp.ru
Russian Federation, Miusskaya pl.4, Moscow, 125047; Moscow, 119991
V. A. Galaktionov
The Keldysh Institute of Applied Mathematics
Email: vova@frolov.pp.ru
Russian Federation, Miusskaya pl.4, Moscow, 125047