Subscribe
pp: 1
Hamid Reza Vaezi Joze , Simon Fraser University, Burnaby
Mark S. Drew , Simon Fraser University, Burnaby
ABSTRACT
Exemplar-based learning has recently gained interest from researchers in a variety of computer science domains because of the prevalence of large amounts of accessible data and storage capacity. Applying the concept of exemplar-based learning to the problem of colour constancy seems odd at first glance since similar nearest neighbour images are not usually affected by precisely similar illuminants and gathering a dataset consisting of scenes for all possible illuminants is indeed impossible. In this paper we instead focus on \em{surfaces} in the image and address the colour constancy problem by unsupervised learning of an appropriate model for each surface in training images. We find models for each test-image surface and finally, illumination estimates are combined. We show state-of-the-art performance compared to current colour constancy algorithms, including when learning based on one image dataset is applied to tests from a different dataset. The proposed method has the advantage of overcoming multi-illuminant situations, which is not possible for most current methods. We test our technique on images with two distinct sources of illumination using a multiple-illuminant colour constancy dataset. The concept proposed here is a completely new approach to the colour constancy problem and provides a simple learning-based framework.
INDEX TERMS
Multiple Illuminants, Color Constancy, Exemplar Based Learning
CITATION
Hamid Reza Vaezi Joze, Mark S. Drew, "Exemplar-Based Colour Constancy and Multiple Illumination", IEEE Transactions on Pattern Analysis & Machine Intelligence, , no. 1, pp. 1, PrePrints PrePrints, doi:10.1109/TPAMI.2013.169