Illumination distribution from brightness in shadows: Adaptive estimation of illumination distribution with unknown reflectance properties in shadow regions

This paper describes a new method for estimating the illumination distribution of a real scene from a radiance distribution inside shadows cast by an object in the scene. First, the illumination distribution of the scene is approximated by discrete sampling of an extended light source. Then the illumination distribution of the scene is estimated from a radiance distribution inside shadows cast by an object of known shape onto another object in the scene. Instead of assuming any particular reflectance properties of the surface inside the shadows, both the illumination distribution of the scene and the reflectance properties of the surface are estimated simultaneously, based on iterative optimization framework. In addition, this paper introduces an adaptive sampling of the illumination distribution of a scene. Rather than using a uniform discretization of the overall illumination distribution, we adaptively increase sampling directions of the illumination distribution based on the estimation at the previous iteration. Using the adaptive sampling framework, we are able to estimate overall illumination more efficiently by using fewer sampling directions. The proposed method is effective for estimating an illumination distribution even under a complex illumination environment