Deep specialized network for illuminant estimation

Abstract

Illuminant estimation to achieve color constancy is an illposed problem. Searching the large hypothesis space for an accurate illuminant estimation is hard due to the ambiguities of unknown reflections and local patch appearances. In this work, we propose a novel Deep Specialized Network (DS-Net) that is adaptive to diverse local regions for estimating robust local illuminants. This is achieved through a new convolutional network architecture with two interacting sub-networks, i.e. an hypotheses network (HypNet) and a selection network (SelNet). In particular, HypNet generates multiple illuminant hypotheses that inherently capture different modes of illuminants with its unique two-branch structure. SelNet then adaptively picks for confident estimations from these plausible hypotheses. Extensive experiments on the two largest color constancy benchmark datasets show that the proposed ‘hypothesis selection’ approach is effective to overcome erroneous estimation. Through the synergy of HypNet and SelNet, our approach outperforms state-of-the-art methods such as [1–3].