Recovery of surface normals and reflectance from different lighting conditions

Abstract

This paper presents a method for finding the surface normals and reflectance of an object from a set of images obtained under different lighting conditions. This set of images, assuming a Lambertian object, can be approximated by a three dimensional linear subspace, under an orthographic camera model and without shadows and specularities. However, a higher dimensional subspace is needed when images present pixels in shadow, specularities or ambient illumination. This paper proposes on the one hand to consider pixels in shadow and specularities as missing data; and on the other hand a rank-four formulation to recover the ambient illumination. An adaptation of the Alternation technique is introduced to compute the sought surface normals and light-source matrices. Experimental results show the good performance of the proposed Alternation-based strategy. © 2008 Springer-Verlag Berlin Heidelberg.