Technical Section: Importance sampling for volumetric illumination of flames

  • Authors:
  • Yingping Zhang;Dengming Zhu;Xianjie Qiu;Zhaoqi Wang

  • Affiliations:
  • Virtual Reality Laboratory, Institute of Computing Technology, Chinese Academy of Sciences, China and Graduate School of the Chinese Academy of Sciences, China;Virtual Reality Laboratory, Institute of Computing Technology, Chinese Academy of Sciences, China;Virtual Reality Laboratory, Institute of Computing Technology, Chinese Academy of Sciences, China;Virtual Reality Laboratory, Institute of Computing Technology, Chinese Academy of Sciences, China

  • Venue:
  • Computers and Graphics
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Physically based rendering of scenes with volumetric illumination of flames remains a challenging problem due to the complexity of their heterogeneous radiative properties. Current bidirectional importance sampling strategies have been focusing on emissive light sources without anisotropic extinction. In this paper, we present an efficient importance sampling method for volumetric light sources with anisotropic extinction. According to the radiative properties of flames, we separate the computation of anisotropic extinction from the evaluation of illumination inside flames and utilize cluster-based hierarchies to rapidly estimate them. To exploit the coherence of radiative voxels, we also propose a new similarity metric to aggregate voxels into clusters. For each pixel to be shaded, we use these clusters to rapidly approximate the importance function of voxels, and draw final illumination samples from clusters. Our results show that this approach substantially reduces the variance of images when rendering scenes with flames.