Estimating the Illuminant Color from the Shading of a Smooth Surface

摘要

The image of a uniform wall illuminated by a spot light often gives a strong impression of the illuminant color. How can it be possible to know if it is a white wall illuminated by yellow light or a yellow wall illuminated by white light. If the wall is a Lambertian reflector, it would not be possible to tell the difference. However, in the real world, some amount of specular reflection is almost always present. In this memo, it is shown that the computation is possible in most practical cases. Light reflection from a surface is usually modeled as having two components: the interface (specular) reflection and the body (diffuse) reflection. For a surface of inhomogeneous material, the spectral composition of the interface reflection component is often similar to that of the illuminant. The problem of computing the illuminant chromaticity from the shading of a single smooth surface is to separate these two components. An image of an illuminated uniform wall, according to the above model, gives only one physical constraint about the illuminant chromaticity, not enough to determine a unique solution. However, since the spatial scale over which the interface reflection changes significantly is much smaller than that of the body reflection, it can be shown that one can effectively exploit the scale difference to find a unique solution, which is often very accurate. The method can also be generalized to compute the illuminant chromaticity for a nonuniform smooth surface. Keywords: Color vision, Shading, Computer generated images, Artificial intelligence.