期刊
IEEE ROBOTICS AND AUTOMATION LETTERS
卷 7, 期 2, 页码 1502-1509出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/LRA.2021.3137888
关键词
Mapping; sensor fusion; multi-robot systems; deep learning for visual perception; data sets for robotic vision
类别
资金
- Jaguar Land Rover [EP/N01300X/1]
- U.K.-EPSRC [EP/N01300X/1]
This paper proposes a novel method for real-time registration of partially overlapping point clouds. The method utilizes an efficient point-wise feature encoder to learn correspondences and a graph-based attention network to refine the matches. The proposed method achieves comparable performance with state-of-the-art methods on different datasets and outperforms existing methods for low overlapping point clouds. Additionally, the method significantly reduces the inference time.
Real-time registration of partially overlapping point clouds has emerging applications in cooperative perception for autonomous vehicles and multi-agent SLAM. The relative translation between point clouds in these applications is higher than in traditional SLAM and odometry applications, which challenges the identification of correspondences and a successful registration. In this paper, we propose a novel registration method for partially overlapping point clouds where correspondences are learned using an efficient point-wise feature encoder, and refined using a graph-based attention network. This attention network exploits geometrical relationships between key points to improve the matching in point clouds with low overlap. At inference time, the relative pose transformation is obtained by robustly fitting the correspondences through sample consensus. The evaluation is performed on the KITTI dataset and a novel synthetic dataset including low-overlapping point clouds with displacements of up to 30 m. The proposed method achieves on-par performance with state-of-the-art methods on the KITTI dataset, and outperforms existing methods for low overlapping point clouds. Additionally, the proposed method achieves significantly faster inference times, as low as 410 ms, between 5 and 35 times faster than competing methods. Our code and dataset are available at https://github.com/eduardohenriquearnold/fastreg.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据