Yasuyuki Matsushita, Stephen Lin, Sing Bing Kang, and Heung-Yeung Shum
We present a method for estimating intrinsic images from a fixed-viewpoint image sequence captured under changing illumination directions. Previous work on this problem reduces the influence of shadows on reectance images, but does not address shading effects which can significantly degrade reectance image estimation under the typically biased sampling of illumination directions. In this paper, we describe how biased illumination sampling leads to biased estimates of reflectance image derivatives. To avoid the effects of illumination bias, we propose a solution that explicitly models spatial and temporal constraints over the image sequence. With this constraint network, our technique minimizes a regularization function that takes advantage of the biased image derivatives to yield reflectanceimages less influenced by shading.
All copyrights reserved by Springer 2004.