Convolutional Color Constancy
This addresses the challenge of accurately inferring illumination color for image processing, with incremental improvements in a specific domain.
The paper tackled the problem of color constancy by reformulating it as a 2D spatial localization task in log-chrominance space, resulting in a nearly 40% performance improvement on standard benchmarks.
Color constancy is the problem of inferring the color of the light that illuminated a scene, usually so that the illumination color can be removed. Because this problem is underconstrained, it is often solved by modeling the statistical regularities of the colors of natural objects and illumination. In contrast, in this paper we reformulate the problem of color constancy as a 2D spatial localization task in a log-chrominance space, thereby allowing us to apply techniques from object detection and structured prediction to the color constancy problem. By directly learning how to discriminate between correctly white-balanced images and poorly white-balanced images, our model is able to improve performance on standard benchmarks by nearly 40%.