Pairwise Point Cloud Registration Using Graph Matching and Rotation-Invariant Features

Registration is a fundamental but critical task in point cloud processing, which usually depends on finding element correspondence from two clouds. However, the of reliable relies establishing robust and discriminative description elements correct matching corresponding elements. In this letter, we develop coarse-to-fine registration strategy, utilizes rotation-invariant features new weighted graph method for iteratively correspondence. method, similarity nodes edges Euclidean feature space are formulated to construct optimization function. The proposed strategy evaluated using benchmark datasets compared with several state-of-the-art methods. Regarding experimental results, our can achieve fine rotation errors less than 0.2 degrees translation 0.1m.