Residual images remove illumination artifacts!

Abstract

In past studies, illumination effects have been proven to cause the most common problems in correspondence algorithms. In this paper, we conduct a study identifying that the residual images (i.e., differences between images and their smoothed versions) contain the important information in an image. We go on to show that this approach removes illumination artifacts between corresponding pairs of images (i.e., optical flow and stereo) using a mixture of synthetic and real-life images. © 2009 Springer Berlin Heidelberg.