Fast Inverse Reflector Design (FIRD)

Abstract

This paper presents a new method ofr a GPU-based computation of outgoing light distribution for inverse reflector design. We propose a fast method to obtain the outgoing light distribution of a parametrized reflector, and compare it with the desired illumination, that works completely in the GPU. We trace millions of rays using a hierarchical height-field representation of the reflector. Multiple reflections are taken into account. The parameters that define the reflector shape are optimized in an iterative procedure in order that the resulting light distribution is as close as possible to the user-provided target light distribution. We show that our method can calculate the reflector lighting at least one order of magnitude faster, even with millions of rays, and complex geometries and light sources.