GLOBAL ILLUMINATION USING LOCAL LINEAR DENSITY-ESTIMATION

This article presents the density estimation framework for generating view-independent global illumination solutions. It works by probabilis tically simulating the light flow in an environment with light particl es that trace random walks originating at luminaires and then using st atistical density estimation techniques to reconstruct the lighting on each surface. By splitting the computation into separate transport an d reconstruction stages, we gain many advantages including reduced mem ory usage, the ability to simulate nondiffuse transport, and natural p arallelism. Solutions to several theoretical and practical difficultie s in implementing this framework are also described. Light sources tha t vary spectrally and directionally are integrated into a spectral par ticle tracer using nonuniform rejection. A new local linear density es timation technique eliminates boundary bias and extends to arbitrary p olygons. A mesh decimation algorithm with perceptual calibration is in troduced to simplify the Gouraud-shaded representation of the solution for interactive display.