Inference of 3-D shape with edge from image brightness

A traditional method to reconstruct a 3-D shape from shading is to structur e an objective functional H(XI, over surface normal distribution X, which i s a weighted average of the regularization terms representing smoothness co nstraints and data term representing the image-irradiance equation; the rec onstruction is then to find the surface normal distribution X, which minimi zes H. However, there is a prominent weakness in that it is difficult to re cover discontinuities in the surface normal at edges. In order to overcome this drawback, we propose a new method of shape from shading by using a con cave-type regularization term for reconstruction of a 3-D shape with edges for non-Lambertian surface. We theoretically prove that the algorithm is co nvergent and effective for recovery of a 3-D shape with edges. We also prov ide experimental comparison of the proposed method with the existing method by using both numerical and real images. Experimental results show that ou r recovery is more accurate.