An approach to robust ICP initialization
This addresses a specific challenge in point cloud registration for applications like robotics or computer vision, but appears incremental as it builds on existing ICP methods.
The paper tackles the problem of initializing the Iterative Closest Point (ICP) algorithm for unlabeled point clouds under rigid transformations by matching covariance ellipsoids and testing principal half-axes matchings, with results including derived robustness bounds and experimental confirmation.
In this note, we propose an approach to initialize the Iterative Closest Point (ICP) algorithm to match unlabelled point clouds related by rigid transformations. The method is based on matching the ellipsoids defined by the points' covariance matrices and then testing the various principal half-axes matchings that differ by elements of a finite reflection group. We derive bounds on the robustness of our approach to noise and numerical experiments confirm our theoretical findings.