Computational models of color perception: characteristics and applications

In this work we present two computational models of color perception: Retinex, the famous model due to Land and McCann and ACE, a new algorithm, recently presented by the author and his colleagues. These two algorithms share a common approach to color appearance estimation but different implementations. The poster introduces the two models focusing on differences and common points. An analysis of their computational characteristics illustrates the way Retinex approach has influenced ACE structure and which aspects differ. Their interesting ability in reproducing human vision properties, like lightness and color constancy suggest their use for several applications. Possible applications are: color appearance estimation, image equalization, detail enhancement, film restoration, automatic interface control, contrast correction, high dynamic range images tone mapping, etc… Exemples are presented and discussed.