Shape recovery from shading by a new neural-based reflectance model

In this paper, we present a neural-based reflectance model of which the phy sical parameters of the reflectivity under different lighting conditions ar e interpreted by the network weights. The idea of our method is to optimize a proper reflectance model by an effective learning algorithm and to recov er the object surface by a simple shape from shading recursive algorithm,vi th this resulting model. Experimental results, including synthetic and real images, were performed to demonstrate the performance of the proposed meth od for practical applications.