A study of incremental update of global illumination algorithms

Abstract

Global illumination solutions provide a very accurate representation of illumination. However, they are usually costly to calculate. In the common case of quasi-static scenario, in which most of the scene is static and only a few objects move, most of the illumination can be reused from previous frames, yielding increased performance. This article studies theoretically the performance of global illumination algorithms for the case of interactive recalculation of quasi-static scenes, concentrating in the Density Estimation on the Tangent Plane algorithm, although the study is applicable to other techniques. The results are validated empirically with a test scene. Guidelines are given to choose the best algorithm for each case.