A survey of iterative closest point algorithm

Registration of 3D point set is a very important problem in computer vision. As a widely used algorithm in 3D point set registration, the ICP algorithm has got much attention of researchers. This paper presents a comprehensive survey of the modified ICP algorithms in recent decade by five stages: selection of registration elements, feature metrics, search strategy, Weight of the Point Pair and the Solution of Rigid Transformation. Finally, this paper compared the standard ICP algorithm and the improved ICP algorithm's registration precision and convergence speed.