Research map / Point clouds / Registration and matching
Efficient and Robust Point Cloud Registration via Heuristics-Guided Parameter Search
Tianyu Huang, Haoang Li, Liangzu Peng, Yinlong Liu, Yunhui Liu
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 |
|---|---|---|---|
| HERE vs parameter-search methods GORE and TR-DE at N=1000, 50-90% outliers | HERE · Simulated Bunny | time: more than 2 orders of magnitude faster | Improved |
| HERE vs GORE with N up to 10000 at 95% outliers | HERE · Simulated Bunny | time: more than 3 orders of magnitude faster than GORE | Improved |
| HERE vs TR-DE with large correspondence sets | HERE · Simulated Bunny | time: 2 orders of magnitude faster than TR-DE | Improved |
| HERE vs SC2-PCR with large correspondence sets | HERE · Simulated Bunny | time: 15 times faster than SC2-PCR | Improved |
| HERE vs TEASER++ with large correspondence sets | HERE · Simulated Bunny | time: 3 times faster than TEASER++ | Improved |
| HERE vs MAC with large correspondence sets | HERE · Simulated Bunny | time: 10 times faster than MAC | Improved |
| HERE vs GORE and SC2-PCR at extreme outlier ratios | HERE · Simulated Bunny (95-99% outliers) | time: more than 10 times faster | Improved |
| increase RANSAC iterations from 1k to 100k | RANSAC · KITTI (FPFH) | registration recall (%): 19.09 -> 95.67 | Improved |
| replace FPFH with FCGF descriptor | RANSAC-1k · KITTI | registration recall (%): 19.09 -> 97.29 | Improved |
| replace FPFH with FCGF descriptor | FGR · KITTI | registration recall (%): 9.73 -> 97.47 | Improved |
| replace FPFH with FCGF descriptor | TEASER++ · KITTI | time (s): 0.36 -> 37.81 | Hurt |
| HERE vs TEASER++ on ETH Trees | HERE · ETH Trees | time (s): 3.79 -> 0.86 | Improved |
Closest papers by meaning
- Deterministic Point Cloud Registration via Novel Transformation Decomposition (CVPR 2022 · similarity 0.92)
- A robust inlier identification algorithm for point cloud registration via $\mathbf{\ell_0}$-minimization (NeurIPS 2024 · similarity 0.92)
- Hunter: Exploring High-Order Consistency for Point Cloud Registration With Severe Outliers (TPAMI 2023 · similarity 0.91)
- MHopReg: Efficient Hierarchical Multi-Hop Graph Search for Point Cloud Registration (CVPR 2026 · similarity 0.91)
- HeMoRa: Unsupervised Heuristic Consensus Sampling for Robust Point Cloud Registration (CVPR 2025 · similarity 0.89)
- Inlier Confidence Calibration for Point Cloud Registration (CVPR 2024 · similarity 0.89)
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.
