Region-Based Image Classification for Automatic Color Correction

Clément Fredembach, Michael Schröder, Sabine Süsstrunk

Abstract: This work presents a framework for image classification based on region information. The main application of this framework is for automatic color correction of real-scene images. Class specific color correction is already implemented in some systems, but image classification usually relies on global image features only. Our region-based approach shows a significant improvement over previously achieved classification for a variety of critical image classes. Improved image classification can directly be translated into improved automatic color correction.