On-line learning of parametric mixture models for light transport simulation
On-line learning of parametric mixture models for light transport simulation
复制标题
DOI：
10.1145/2601097.2601203
复制
发表时间：
2014-07
期刊：
影响因子：
--
通讯作者：
J. Vorba;Ondrej Karlik;M. Šik;Tobias Ritschel;Jaroslav Křivánek
中科院分区：
文献类型：
--
作者：
J. Vorba;Ondrej Karlik;M. Šik;Tobias Ritschel;Jaroslav Křivánek
用于光传输模拟的蒙特卡罗技术在构造光传输路径时依赖于重要性采样。先前的工作已经表明，合适的采样分布可以从渲染之前分布在场景中的粒子中恢复。我们建议用一个参数混合模型来表示分布，该模型是从一个潜在的无限粒子流中以在线（即渐进）的方式训练的。这可以在具有复杂照明的场景中恢复良好的采样分布，其中所需的粒子数量可能超过可用内存。使用这些分布对散射方向和光发射进行采样，可以显著提高处理复杂照明时最先进的光传输模拟算法的性能。
Monte Carlo techniques for light transport simulation rely on importance sampling when constructing light transport paths. Previous work has shown that suitable sampling distributions can be recovered from particles distributed in the scene prior to rendering. We propose to represent the distributions by a parametric mixture model trained in an on-line (i.e. progressive) manner from a potentially infinite stream of particles. This enables recovering good sampling distributions in scenes with complex lighting, where the necessary number of particles may exceed available memory. Using these distributions for sampling scattering directions and light emission significantly improves the performance of state-of-the-art light transport simulation algorithms when dealing with complex lighting.