Research map / Point clouds / Registration and matching
HRegNet: A Hierarchical Network for Large-Scale Outdoor LiDAR Point Cloud Registration
Fan Lu, Guang Chen, Yinlong Liu, Lijun Zhang, Sanqing Qu, Shu Liu, Rongqi Gu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add fine registration in layer 2 on top of coarse layer 3 (ablation) | HRegNet · KITTI odometry | RRE (deg): 0.55 -> 0.29 | Improved |
| add fine registration in layer 2 on top of coarse layer 3 (ablation) | HRegNet · KITTI odometry | RTE (m): 0.16 -> 0.15 | Improved |
| add fine registration in layer 1 on top of layer 2 (ablation) | HRegNet · KITTI odometry | RTE (m): 0.15 -> 0.12 | Improved |
| add fine registration in layer 1 on top of layer 2 (ablation) | HRegNet · KITTI odometry | registration recall: 99.2% -> 99.7% | Improved |
| add fine registration in layer 1 on top of layer 2 (ablation) | HRegNet · KITTI odometry | RRE (deg): 0.29 -> 0.29 | No clear effect |
| add fine registration in layers 2 and 1 (inference time cost) (ablation) | HRegNet · KITTI odometry | time (ms): 96.9 -> 106.2 | Hurt |
| add original similarity features F_S^O (ablation) | HRegNet · KITTI odometry | registration recall: 99.1% -> 99.7% | Improved |
| add neighbor-aware similarity features F_S^N (ablation) | HRegNet · KITTI odometry | registration recall: 99.4% -> 99.7% | Improved |
| add both similarity features F_S (ablation) | HRegNet · KITTI odometry | registration recall: 98.7% -> 99.7% | Improved |
| add both similarity features F_S (ablation) | HRegNet · KITTI odometry | RTE (m): 0.19 -> 0.12 | Improved |
| add both similarity features F_S (ablation) | HRegNet · KITTI odometry | RRE (deg): 0.46 -> 0.29 | Improved |
| add both similarity features F_S (inference time cost) (ablation) | HRegNet · KITTI odometry | time (ms): 88.0 -> 106.2 | Hurt |
Later papers that compared with this method
2 later accepted papers report HRegNet as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- RegFormer: An Efficient Projection-Aware Transformer Network for Large-Scale Point Cloud Registration (ICCV 2023 · better on 3 of 5 shared results)
- SuperLine3D: Self-Supervised Line Segmentation and Description for LiDAR Point Cloud (ECCV 2022 · better on 3 of 3 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- HRegNet: A Hierarchical Network for Efficient and Accurate Outdoor LiDAR Point Cloud Registration (TPAMI 2023 · similarity 0.99)
- GraphI2P: Image-to-Point Cloud Registration with Exploring Pattern of Correspondence via Graph Learning (CVPR 2025 · similarity 0.88)
- Unlocking Generalization Power in LiDAR Point Cloud Registration (CVPR 2025 · similarity 0.87)
- A Consistency-Aware Spot-Guided Transformer for Versatile and Hierarchical Point Cloud Registration (NeurIPS 2024 · similarity 0.87)
- FreeReg: Image-to-Point Cloud Registration Leveraging Pretrained Diffusion Models and Monocular Depth Estimators (ICLR 2024 · similarity 0.87)
- 3D Focusing-and-Matching Network for Multi-Instance Point Cloud Registration (NeurIPS 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
