Fast Global Point Cloud Registration using Semantic NDT
Robust and accurate point cloud registration is an essential part of many robotic tasks such as SLAM or object pose retrieval. This paper presents a novel approach for global 3D point cloud registration that combines the normal distributions transform and oriented point pair framework and introduces the NDT distance histogram to quickly generate and test candidate transforms. The method optionally leverages semantic information for greater speed and demonstrates superior run-time and compute efficiency compared to state-of-the-art approaches.
- Published in:
Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) - Type:
Inproceedings - Authors:
- Year:
2024
Citation information
: Fast Global Point Cloud Registration using Semantic NDT, Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024, Schirmer.etal.2024a,
@Inproceedings{Schirmer.etal.2024a,
author={Schirmer, Robert; Vaskevicius, Narunas; Biber, Peter; Stachniss, Cyrill},
title={Fast Global Point Cloud Registration using Semantic NDT},
booktitle={Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
year={2024},
abstract={Robust and accurate point cloud registration is an essential part of many robotic tasks such as SLAM or object pose retrieval. This paper presents a novel approach for global 3D point cloud registration that combines the normal distributions transform and oriented point pair framework and introduces the NDT distance histogram to quickly generate and test candidate transforms. The method...}}