Double hierarchies for efficient sampling in Monte Carlo rendering

We propose a novel representation of the light field tailored to improve importance sampling for Monte Carlo rendering. The domain of the light field i.e., the product space of spatial positions and directions is hierarchically subdivided into subsets on which local models characterize the light transport. The data structure is based on double trees, and only approximates the exact light field, but enables efficient queries for importance sampling and easy setup by tracing photons in the scene. The framework is simple yet flexible, supports any type of local model for representing the light field, provided it can be efficiently importance sampled, and progressive refinement with an arbitrary number of photons. Last, we provide a reference open source implementation of our method.