Illumination driven plant growth based on ray density estimation

Abstract

Light interaction is one of the most important factors in developing realistic plant models. Plants react to received illumination by bending branches, adapting their growth rate, orienting leaves and flowers, producing larger or smaller leaves, etc. In this paper, we present a novel approach to simulate plant growth as a response to environment illumination. The basic idea of our algorithm is to simulate light transport in the environment in which plants grow by tracing light particles originating from light sources. Both intensity and mean direction of incident illumination are determined easily. This is based on a ray density estimation of the environment illumination and the calculation of a predominant illumination direction. An adaptive spatial data structure is used to store the rays along which light particles travel in space. This data structure allows efficient calculation of ray density at locations where the algorithm needs to query incident illumination. Our approach is influenced by photon mapping. As a result, it incorporates direct and indirect illumination, is both flexible and accurate, is easy to implement, and supports arbitrary global illumination. It also provides a way to decouple the illumination querying from the plant geometry. This allows for queries at any position within the environment, as opposed to a particle based approach, which needs an actual surface to estimate energy density (= incident illumination). Furthermore, using a non-uniform, adaptive data structure for storing the rays, calculation time and storage requirements are kept within reasonable limits.