Three-dimensional modeling from two-dimensional video based on neural network

摘要

In this paper, we present a new approach for determining the reflectance properties of surface and recovering 3D shapes from intensity images. The proposed approach is based on using the neural networks as a parametric representation of the three-dimensional object and the shape from shading problem is formulated as the minimization of an intensity error function with respect to the network weights. The estimated reflectance parameters provide the range data with intensity distributions. Therefore, we generate three reference images of a range sphere, which has the same diameter as that of the sample, from the same viewpoint but with different light directions. The new algorithms for data driven, stable, update the surface slope and height maps are proposed. This approach significantly reduce the residual errors. In comparison with the traditional methods. Some experimental results demonstrating that this method improves shape accuracy are shown.