Bayesian Color Constancy Revisited
Computational color constancy is the task of estimating the true reﬂectances of visible surfaces in an image. In this paper we follow a line of research that assumes uniform illumination of a scene, and that the principal step in estimating reﬂectances is the estimation of the scene illuminant. We review recent approaches to illuminant estimation, ﬁrstly those based on formulae for normalisation of the reﬂectance distribution in an image — so-called grey-world algorithms, and those based on a Bayesian formulation of image formation. In evaluating these previous approaches we introduce a new tool in the form of a database of 568 high-quality, indoor and outdoor images, accurately labelled with illuminant, and preserved in their raw form, free of correction or normalisation. This has enabled us to establish several properties experimentally. Firstly automatic selection of grey-world algorithms according to image properties is not nearly so effective as has been thought. Secondly, it is shown that Bayesian illuminant estimation is signiﬁcantly improved by the improved accuracy of priors for illuminant and reﬂectance that are obtained from the new dataset.