Abstract

This paper presents a new importance sampling strategy for Monte Carlo ray tracing in which a rough estimate of the irradiance based on the photon map is combined with the local reflection model to construct more efficient probability density functions that can be used in an importance sampling scheme.
The algorithm gives unbiased results, handles arbitrary reflection models and it is particularly efficient in scenes with highly non-uniform indirect illumination. Initial results and comparisons with traditional importance sampling strategies indicate a reduction in the noise level of more than 70%

Key Words: Global Illumination, Path Tracing, Importance Sampling, Photon Map

Reference:

Henrik Wann Jensen: "Importance Driven Path Tracing using the Photon Map". In Rendering Techniques '95. Eds. P.M. Hanranhan and W. Purgathofer. Springer-Verlag. pp. 326-335, 1995

Click here to download a 1 MB gzip'edpostscript version of the paper

The Colour Images in the Paper