Abstract: A framework for the recovery of smooth surface shapes from shading images with unknown light source is presented.A single hidden layer wavelet neural network is used to approximate the complex nonlinear reflectance map function and reconstruct the shape as minimizing an error function over the network weight. Object surface is recovered with recursive algorithm and hierarchical implementation is employed to reduce the computation.Under the proposed model,the lighting parameters are not necessary. Experimental results demonstrate the new algorithm is more robust and accurate than traditional ones.