This paper introduces a simple idea of RGB to spectral image conversion without any expensive multi-spectral devices. A spectral reflectance with the closest colorimetric value to that of RGB pixel is picked up from the spectral palette and embedded in each pixel of RGB image. SVD (Singular Value Decomposition) compresses the high-resolution spectral image to a compact data. A conventional RGB camera image is transformed into a pseudo-spectral image and facial skin color appearance is estimated under different illuminants. The paper reports a facial color reproduction experiments using skin color palette extracted from JIS spectral data base SOCS and discusses the color accuracy after SVD compression with low dimensional parameters.
Sousuke Kagaya, Takahiko Horiuchi, Hiroaki Kotera, "Examination on Pseudo-Spectral Color Reproduction of Facial Image Using Spectral Skin Color Palette" in Proc. IS&T Int'l Conf. on Digital Printing Technologies (NIP22), 2006, pp 382 - 385, https://doi.org/10.2352/ISSN.2169-4451.2006.22.1.art00017_2