Shape from shading (SFS) is a key technique in the domain of computer vision, which is to recovery the three-dimensional (3D) shape of an object surface from a single shading image. SFS techniques have wide applications. We introduce the development status of SFS methods in recent years and divide these methods into four groups: propagation and partial differential equation (PDE) methods, optimization methods, local methods and linear methods. At the same time, the characteristics of various SFS algorithms is analysed. The further development trend of the SFS methods is also prospected.%从明暗恢复形状(SFS)是计算机视觉中三维重构问题的关键技术之一,其原理是利用单幅图像的明暗变化来恢复物体表面的三维形状,应用领域十分广泛.介绍了近些年来SFS方法的发展现状,将SFS算法划分为四种类型:演化和偏微分方程(PDE)方法、最优化方法、局部化方法以及线性化方法.同时,分析了各种SFS算法的特点,并对SFS技术的进一步发展趋势进行了展望.
展开▼
