提出LEADER框架，提升激光雷达重定位在复杂场景下的精度与鲁棒性。
LEADER: Learning Reliable Local-to-Global Correspondences for LiDAR Relocalization
- 设计几何编码器捕捉多尺度空间特征，增强点云表征能力。
- 引入截断相对可靠性损失，降低噪声点对结果的干扰。
- 在Oxford RobotCar和NCLT数据集上误差分别降低24.1%和73.9%，适合高精度定位应用。
激光雷达重定位因可在复杂三维环境中实现精确的6自由度位姿估计而受到关注。基于学习的回归方法通过直接预测全局位姿，无需显式地图存储，效率较高。然而，这些方法常因对所有预测点一视同仁，在挑战性场景中易受噪声和异常值影响。本文提出LEADER，一种基于简单而有效的几何编码器的鲁棒激光雷达重定位框架。首先，提出基于投影的鲁棒几何编码器架构，以捕捉多尺度几何特征，提升几何表示能力；随后，设计截断相对可靠性损失，建模点级不确定性，减轻不可靠预测的影响。在Oxford RobotCar和NCLT数据集上的大量实验表明，LEADER优于现有最先进方法，分别实现24.1%和73.9%的位置误差相对降低。源代码已发布于https://github.com/JiansW/LEADER。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。