Selective application of Metropolis Light Transport for hard sampling illumination phenomena
Abstract:
In this paper a method for combining Ordinary Monte Carlo (OMC) and Markov Chain Monte Carlo (MCMC) is proposed. We combine Bidirectional Path Tracing (BPT) and Multiplexed Metropolis Light Transport (MMLT) in image space by using per pixel mask that is obtained from BPT noise analysis. The resulting method is more accurate than BPT and produces more pleasant image than MMLT.
