Hideki HAYAKAWA
Shape, surface reflectance, light intensity and
light direction from shading images
Abstract:We propose a method for estimating the surface normal and surface reflectance of objects without a priori knowledge of the light intensities and light directions. First, we construct a pxf image data matrix I from p pixel image intensity data through f frames by moving a light source arbitrarily. Using the Lambertian assumption, the image data matrix I can be written as the product of two matrices S and L, S representing the surface normal and surface reflectance, and L representing the light intensities and light directions. Based on this formulation, we show that the image data matrix I is of rank 3. Based on the observation, we use a singular value decomposition technique and useful constraints to factorize the image data matrix. Our method can also treat both self-shadow and cast-shadow without assumptions. We demonstrate the effectiveness of our method through performance analysis, laboratory experiments on Lambertian reflectance objects and hybrid reflectance objects, and an out-of-laboratory experiment on a miniature model in a lobby.