Shape distortion analysis of the shape-from-shading algorithm using Jacobi iterative method

In recent research on shape-from-shading much focus has been put on the minimization approaches and on the Tsai and Shah linear approach. We presented a novel non-linear approach using the Jacobi iterative method, which is as simple in implementation as the linear method and as accurate as those minimization ones. This paper first, analyzes our method to clarify how the shape distortion is generated. Then, the effects of the bi-directional estimation of the shape on reducing the distortions are examined with numerical examples.