LEADER: Learning Reliable Local-to-Global Correspondences for LiDAR Relocalization
For autonomous systems requiring accurate 6-DoF pose estimation in complex 3D environments, LEADER provides a more robust learning-based solution.
LEADER introduces a robust LiDAR relocalization framework using a geometric encoder and a truncated relative reliability loss to handle noise and outliers, achieving 24.1% and 73.9% reductions in position error on Oxford RobotCar and NCLT datasets, respectively.
LiDAR relocalization has attracted increasing attention as it can deliver accurate 6-DoF pose estimation in complex 3D environments. Recent learning-based regression methods offer efficient solutions by directly predicting global poses without the need for explicit map storage. However, these methods often struggle in challenging scenes due to their equal treatment of all predicted points, which is vulnerable to noise and outliers. In this paper, we propose LEADER, a robust LiDAR-based relocalization framework enhanced by a simple, yet effective geometric encoder. Specifically, a Robust Projection-based Geometric Encoder architecture which captures multi-scale geometric features is first presented to enhance descriptiveness in geometric representation. A Truncated Relative Reliability loss is then formulated to model point-wise ambiguity and mitigate the influence of unreliable predictions. Extensive experiments on the Oxford RobotCar and NCLT datasets demonstrate that LEADER outperforms state-of-the-art methods, achieving 24.1% and 73.9% relative reductions in position error over existing techniques, respectively. The source code is released on https://github.com/JiansW/LEADER.