An approach to robust ICP initialization
2022-12-10Code Available0· sign in to hype
Alexander Kolpakov, Michael Werman
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/sashakolpakov/icp-initOfficialIn papernone★ 7
Abstract
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.