Abstract: Color constancy refers to the ability of the human visual system to stabilize
the color appearance of surfaces under an illuminant change. In this work we studied how the interrelations among nine colors are perceived under illuminant changes, particularly whether they remain stable across 10 different conditions (5 illuminants and 2 backgrounds). To do so we have used a paradigm that measures several colors under an immersive state of adaptation. From our measures we defined a perceptual structure descriptor that is up to 87% stable over all conditions, suggesting that color category features could be used to predict color constancy. This is in agreement with previous results on the stability of border categories [1,2] and with computational color constancy
algorithms [3] for estimating the scene illuminant.

Abstract: Automatic document content processing is affected by artifacts caused by the shape
of the paper, non-uniform and diverse color of lighting conditions. Fully-supervised
methods on real data are impossible due to the large amount of data needed. Hence, the
current state of the art deep learning models are trained on fully or partially synthetic images. However, document shadow or shading removal results still suffer because: (a) prior methods rely on uniformity of local color statistics, which limit their application on real-scenarios with complex document shapes and textures and; (b) synthetic or hybrid datasets with non-realistic, simulated lighting conditions are used to train the models. In this paper we tackle these problems with our two main contributions. First, a physically constrained learning-based method that directly estimates document reflectance based on intrinsic image formation which generalizes to challenging illumination conditions. Second, a new dataset that clearly improves previous synthetic ones, by adding a large range of realistic shading and diverse multi-illuminant conditions, uniquely customized to deal with documents in-the-wild. The proposed architecture works in two steps. First, a white balancing module neutralizes the color of the illumination on the input image. Based on the proposed multi-illuminant dataset we achieve a good white-balancing in really difficult conditions. Second, the shading separation module accurately disentangles the shading and paper material in a self-supervised manner where only the synthetic texture is used as a weak training signal (obviating the need for very costly ground truth with disentangled versions of shading and reflectance). The proposed approach leads to significant generalization of document reflectance estimation in real scenes with challenging illumination. We extensively evaluate on the real benchmark datasets available for intrinsic image decomposition and document shadow removal tasks. Our reflectance estimation scheme, when used as a pre-processing step of an OCR pipeline, shows a 21% improvement of character error rate (CER), thus, proving the practical applicability. The data and code will be available at: https://github.com/cvlab-stonybrook/DocIIW.