提出了一种新的物体三维形状恢复的遗传算法方法。用固定位置的摄像机在不同位置的球扩展光源下获取图象序列，并用遗传算法对物体表面每一点的法线矢量进行快速搜索。实验结果表明，此方法能有效地恢复扩展光源下物体的三维形状。不仅放宽了对光源和物体表面的限制，而且精度及鲁棒性均有很大提高。

In this paper, we propose a novel method to recover the 3D shape of objects from two image sequences. At first, for eliminating the limitations of the conventional shape from shading methods, we develop a spherical light source model. Based on this model, a fast genetic algorithm is then used for the shape and reflectance recovery from the diffuse reflection component extracted from a pair of image sequences. An image sequence is acquired with each image taken at a different light position. As a result, the normal of a surface patch which is not in the light source moving plane can also be estimated robustly. Our method is verified by both real and simulated experiments.