This paper proposes an image-dependent approach to interchange a color between different objects in different scenes. An unpleasant scene color is automatically corrected to match to that of reference pleasant scene. The paper introduces a “scene-to-scene color transfer model”, based on Histogram Rescaling between color gamuts. The Histogram Rescaling method can easily adapt an image gamut to an objective target image gamut by simply setting the lowest and highest source values corresponding to the end points of the destination. First the lightness rescaling is performed and next the chroma is rescaled. After the histogram rescaling of lightness L, the chroma components are divided into 16 segments by ΔH at hue angle H. Where source and target image center points on the CIELAB color space are separately set. Then source image's chroma C of each segment is expand or compressed by histogram rescaling referencing target image's chroma. Finally a center of source image is transferred into a center of target image and scene color interchange is performed. Proposed system is applied to preferred color reproduction, scene simulation, industrial design, etc. Experiments on total scene color transfer and local object color interchange are presented, where a color atmosphere is transferred from one scene to another.
Ryoichi Saito, Takahiko Horiuchi, Hiroaki Kotera, "Scene Color Interchange using Histogram Rescaling" in Proc. IS&T Int'l Conf. on Digital Printing Technologies (NIP22), 2006, pp 378 - 381, https://doi.org/10.2352/ISSN.2169-4451.2006.22.1.art00016_2