A Novel Monte Carlo Noise Reduction Operator

Monte Carlo noise appears as outliers and as interpixel incoherence in a typical image rendered at low sampling density. Unfortunately, none of the previous approaches can reduce both types of noise in a unified way. In this article, we propose such a unified Monte Carlo noise reduction approach using bilateral filtering. We extended the standard bilateral filtering method and built a new local adaptive noise reduction kernel. The new operator suppresses the outliers and interpixel incoherence in a noniterative way.