Improvement of Accuracy for Gaussian Curvature Using Modification Neural Network

摘要

This paper proposes a new approach to recover the relative magnitude of Gaussian curvature of the test object from four shading images using modified neural network. The method is expanded to an object with color texture using four shading images taken under the different light source directions. Neural network mapps four image irradiances on the test object onto a point on a sphere. The area value surrounded by four mapped points onto a sphere gives an approximate value of Gaussian curvature. To get more accurate Gaussian curvature, the modification neural network is introduced and learned for the synthesized 2-D basis function consisting of 2-D cosine function. It is shown that learnt NN gives better accuracy for the relative magnitude of Gaussian curvature of the test object.