Title :
Estimation of the size and location of multiple area light sources
Author :
Zhou, Wei ; Kambhamettu, Chandra
Author_Institution :
Dept. of Comput. & Inf. Sci., Delaware Univ., Newark, DE, USA
Abstract :
Most illuminant estimation algorithms worked on point light sources or directional light sources. Little attempt has been made, however, to estimate area light sources. In this paper, we present a novel scheme that estimates the size and location of multiple area light sources using a set of stereo images of a sphere with a shiny surface. The parameters of the area light source are estimated by a novel algorithm which minimizes the matching error between the corresponding specular patches. Experiments on real images show that our method is accurate and robust in estimating the parameters of the area light source.
Keywords :
light sources; minimisation; parameter estimation; stereo image processing; directional light sources; illuminant estimation algorithms; light source location estimation; light source size estimation; matching error minimization; multiple area light source estimation; parameter estimation; point light source; sphere stereo images; Application software; Computer graphics; Layout; Light sources; Object detection; Parameter estimation; Rendering (computer graphics); Robustness; Shape; Stereo vision;
Conference_Titel :
Pattern Recognition, 2004. ICPR 2004. Proceedings of the 17th International Conference on
Print_ISBN :
0-7695-2128-2
DOI :
10.1109/ICPR.2004.1334506