Efficient potential equation solutions for global illumination computation

Abstract

The potential equation provides a natural mathematical basis for all the illumination computation methods based on the strategy of shooting light. The particle tracing technique that physically simulates the particle model of light, can also be considered a general, random walk-based solution to the potential equation. As the complexity of the environment increases, the straightforward particle tracing method becomes inefficient. In this paper we discuss different biasing methods that could be used to increase the efficiency of the particle tracing process. In particular we present an importance-driven scheme in which we use an approximate potential value as the biasing function for directing a majority of the random walks through regions of importance in the environment, thus efficiently estimating the light flux in these regions. A straightforward simple implementation of this scheme has been carried out and the results that we have obtained show that the scheme is very promising.