A Fast Multi-Resolution Iterative Closest Point Algorithm

Iterative Closest Point (ICP) is a classical algorithm for rigid point set registration. However, with the number of points in the set increasing, its computational efficiency usually suffers a reduction, which limits the practical applications of this algorithm. Based on Weber's Law in psychophysics, this paper proposes a fast ICP algorithm based on hierarchical and multi-resolution model for point set matching. In the proposed algorithm, an initial transformation matrix between two point sets acquired firstly through the fast iterative computation at lower resolution, and then it is used as the initial value for a more precise registration at high resolution and we complete the overall rigid point set registration process. By this proposal, the time cost for computing the initial value at high resolution could be saved and thus the efficiency of the overall ICP registration algorithm would be greatly improved. Experimental results show that the proposed algorithm is apparently more efficient than the traditional ICP algorithm on both 2D and 3D point sets.