Multi-band Photometric Stereo Using Random Sampling of Channels and Pixels

Daisuke Miyazaki, Koumei Hamaen

Abstract: One of the main problems faced by the photometric stereo method is that several measurements are required, as this method needs illumination from light sources from different directions. A solution to this problem is the color photometric stereo method, which conducts one-shot measurements by simultaneously illuminating lights of different wavelengths. However, the classic color photometric stereo method only allows measurements of white objects, while a surface-normal estimation of a multicolored object using this method is theoretically impossible. This paper estimates the surface normal of a multi-colored object under multi-spectral lighting. This is a difficult problem since the albedo is different for each pixel and each channel. We solve this problem by sampling some pixels randomly. If we randomly sample neighboring pixels, the probability of picking the same albedo pixels may be high. Therefore, if the sampled pixels all have the same albedo, we can determine the surface normal uniquely. To demonstrate the effectiveness of this study, a measurement device with seven colors is used.