Photorealistic animation rendering with population Monte Carlo energy redistribution

The requirement of realism in animation rendering has increased steadily with advances of computer hardware and algorithms. To generate photorealistic animation requires the estimation of a large number of temporally and spatially correlated path integrals. In this dissertation a space-time algorithm is presented that exploits the spatial and temporal coherence among all path integrals based on the Markov chain Monte Carlo (MCMC) framework when rendering a physically-correct animation. Although sample reuse can improve rendering efficiency, the choice of good mutation strategies for MCMC algorithms is non-trivial and has a major impact on image quality.;We adapt the population Monte Carlo framework into an energy redistribution algorithm to reuse information from important samples with a mutation process whose mutation strategy is adapted "on-the-fly." The proposed algorithm is called population Monte Carlo energy redistribution (PMC-ER) and is self-tuning to a large extent. A new lens perturbation technique is presented that simplifies the computation and control of caustics perturbation. This increases the perturbation success rate. In addition, two path-regeneration methods are described that aim to concentrate more computation on "high perceptual variance" regions and "hard-to-find-but-important" paths. Additionally, a new perturbation technique called time perturbation is developed to explore the temporal coherence among paths.;In order to make animation rendering feasible, we develop a parallel rendering system to distribute the iterative computational tasks to a pool of computers. Each task is rendered with a set of task-based parameters without introducing bias. The resulting animations are perceptually better than those rendered frame-by-frame. In this dissertation we demonstrate that population Monte Carlo energy redistribution can enhance rendering efficiency i.e. rendering speed and quality by automatically adjusting the rendering parameters when exploiting the temporal and spatial coherence among light transport paths and by concentrating more computation on visually important regions on the image plane and in the path space.