Cascading Convolutional Color Constancy
This work addresses the problem of color constancy for computer vision applications, offering an incremental improvement in robustness and generalization across different cameras and scenes.
The paper tackles the challenge of robust illumination estimation in computational color constancy by proposing the Cascading Convolutional Color Constancy (C4) method, which achieves superior performance on public benchmarks like Color Checker and NUS 8-Camera compared to state-of-the-art methods, particularly in difficult scenes.
Regressing the illumination of a scene from the representations of object appearances is popularly adopted in computational color constancy. However, it's still challenging due to intrinsic appearance and label ambiguities caused by unknown illuminants, diverse reflection property of materials and extrinsic imaging factors (such as different camera sensors). In this paper, we introduce a novel algorithm by Cascading Convolutional Color Constancy (in short, C4) to improve robustness of regression learning and achieve stable generalization capability across datasets (different cameras and scenes) in a unique framework. The proposed C4 method ensembles a series of dependent illumination hypotheses from each cascade stage via introducing a weighted multiply-accumulate loss function, which can inherently capture different modes of illuminations and explicitly enforce coarse-to-fine network optimization. Experimental results on the public Color Checker and NUS 8-Camera benchmarks demonstrate superior performance of the proposed algorithm in comparison with the state-of-the-art methods, especially for more difficult scenes.