Decomposition of reflection and scattering by multiple-weighted measurements

Tsuyoshi Takatani, Yasuhiro Mukaigawa, Yasuyuki Matsushita, Yasushi Yagi

Abstract: An observed image is composed of multiple components based on optical phenomena, such as light reflection and scattering. Decomposing the observed image into individual components is an important process for various computer vision tasks. No general approach to combine them exists although many decomposition methods exist. This paper proposes a general approach to combine different decomposition methods in a linear algebraic manner called multiple-weighted measurements.Experimental results show that the proposed approach decomposes observed images into four optical components based on diffuse and specular reflection and single and multiple scattering. The decomposed components are applied to material segmentation as an application.