Global illumination using local linear density estimation

This paper presents the density estimation framework for generating view�independent global illumination solutions. It works by probabilis� ticly simulating the light �ow in an environment with light particles that trace random walks originating at luminaires and then using statistical density estimation techniques to reconstruct the lighting on each surface. By splitting the computation into separate transport and reconstruction stages � we gain many advantages including reduced memory usage � the ability to simulate non�di�use transport � and natural parallelism. This paper also describes how several theoretical and practical di�cul� ties can be overcome in implementing this framework. Light sources that vary spectrally and directionally are integrated into a spectral particle tracer using non�uniform rejection. A new local linear density estimation technique eliminates boundary bias and extends to arbitrary polygons. A mesh decimation algorithm with perceptual calibration is introduced to simplify the Gouraud�shaded representation of the solution for interactive display. CR Categories and Subject Descriptors � I.3.0 �Computer Graphics�� General � I.3.6 �Computer Graphics� � Methodology and Techniques. Additional Key Words and Phrases � realistic image synthesis � densit