Research map / Point clouds / Registration and matching
Robust inlier selection for registration
Reject outlier correspondences using geometric transformers, inlier evaluation and randomized solvers.
29 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2019–2026) · an approach in Registration and matching, keyword cluster “cloud registration · overlap · feature”
State History Benchmarks Tried Checklist Preprints Who Export
Where it stands
Still moving The best reported result is still improving on 1 of 2 tracked benchmarks.
7 papers in 2023–24 and 9 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic held steady, from 20.0% to 19.6%.
Open questions the data points to
- 3DMatch, FPFH (RR (%)): best went from 83.98 (up to 2024) to 88.48 in 2025 — Progressive Correspondence Regenerator for Robust 3D Registration (CVPR 2025 · 5 citations)
What looks settled
- 3DMatch, FCGF (RR (%)): best up to 2024 93.28, best in 2025–26 93.96 from 2 papers, less than 1% better.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2019 · SDRSAC: Semidefinite-Based Randomized Approach for Robust Point Cloud Registration Without Correspondences (CVPR 2019 · 104 citations)
- 2020 · Feature-Metric Registration: A Fast Semi-Supervised Approach for Robust Point Cloud Registration Without Correspondences (CVPR 2020 · 325 citations)
- 2021 · PointDSC: Robust Point Cloud Registration Using Deep Spatial Consistency (CVPR 2021 · 359 citations)
- 2022 · Geometric Transformer for Fast and Robust Point Cloud Registration (CVPR 2022 · 501 citations)
- 2023 · BUFFER: Balancing Accuracy, Efficiency, and Generalizability in Point Cloud Registration (CVPR 2023 · 102 citations)
- 2024 · Dynamic Cues-Assisted Transformer for Robust Point Cloud Registration (CVPR 2024 · 25 citations)
- 2025 · HeMoRa: Unsupervised Heuristic Consensus Sampling for Robust Point Cloud Registration (CVPR 2025 · 8 citations)
- 2026 · Rethinking 2D-3D Registration: A Novel Network for High-Value Zone Selection and Representation Consistency Alignment (CVPR 2026 · 2 citations)
Earliest papers here
- SDRSAC: Semidefinite-Based Randomized Approach for Robust Point Cloud Registration Without Correspondences (CVPR 2019 · 104 citations)
- Robust Point Cloud Based Reconstruction of Large-Scale Outdoor Scenes (CVPR 2019 · 18 citations)
- Feature-Metric Registration: A Fast Semi-Supervised Approach for Robust Point Cloud Registration Without Correspondences (CVPR 2020 · 325 citations)
Benchmark records
The best number each year that beat every earlier one, among papers of this approach reporting their own method. Numbers are extracted by AI from each paper's text; check the paper before citing.
3DMatch, FCGF · RR (%) (higher is better)
4 papers, 2022–2025 · small gains only
| Year | Best so far | Set by |
|---|---|---|
| 2022 | 93.28 | SC2-PCR: A Second Order Spatial Compatibility for Efficient and Robust Point Cloud Registration (CVPR 2022 · 181 citations) |
| 2025 | 93.96 | Progressive Correspondence Regenerator for Robust 3D Registration (CVPR 2025 · 5 citations) |
3DMatch, FPFH · RR (%) (higher is better)
4 papers, 2022–2025 · still improving
| Year | Best so far | Set by |
|---|---|---|
| 2022 | 83.98 | SC2-PCR: A Second Order Spatial Compatibility for Efficient and Robust Point Cloud Registration (CVPR 2022 · 181 citations) |
| 2025 | 88.48 | Progressive Correspondence Regenerator for Robust 3D Registration (CVPR 2025 · 5 citations) |
What has been tried
23 papers here report 230 measured changes (a component added, removed or swapped, compared with a baseline): helped 77%, no clear effect 6%, hurt 18%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| add dynamic updating with residual connection | 3DMatch · RR (%) | 91.0 -> 92.1 | Helped | Dynamic Cues-Assisted Transformer for Robust Point Cloud Registration (CVPR 2024 · 25 citations) |
| add transformation-invariant geometric constraints for inlier confidence calibration | ModelNet40 · MAE(R) | 1.0123 -> 0.0012 | Helped | Inlier Confidence Calibration for Point Cloud Registration (CVPR 2024 · 24 citations) |
| Replace sampled-correspondence indicator Iq-V1 with consistency-aware Iq-V2 reward propagation | 3DMatch (FPFH) · RR (%) | 77.33% -> 78.13% | Helped | HeMoRa: Unsupervised Heuristic Consensus Sampling for Robust Point Cloud Registration (CVPR 2025 · 8 citations) |
| Add spot-guided cross-attention and consistency-aware self-attention | 3DMatch · RR (%) | 94.45 -> 96.01 | Helped | A Consistency-Aware Spot-Guided Transformer for Versatile and Hierarchical Point Cloud Registration (NeurIPS 2024 · 8 citations) |
| use the density-robust encoder instead of KPConv | 3DCSR Kinect-lidar · RR | 57.7 -> 66.7 | Helped | Cross-PCR: A Robust Cross-Source Point Cloud Registration Framework (AAAI 2025 · 6 citations) |
| progressive regeneration instead of one-stage processing | 3DMatch · RR | 86.81 -> 88.48 | Helped | Progressive Correspondence Regenerator for Robust 3D Registration (CVPR 2025 · 5 citations) |
| Add overlap-aware Sinkhorn Distance, including overlap filtering and overlap-aware initialization | 3DMatch, 5000 samples · RR (%) | 90.4 -> 93.5 | Helped | Multi-scale Consistency for Robust 3D Registration via Hierarchical Sinkhorn Tree (NeurIPS 2024 · 2 citations) |
| Use proposed registration instead of FPFH+RANSAC | Real3D-AD · P-AUROC | 72.6 -> 87.8 | Helped | Registration is a Powerful Rotation-Invariance Learner for 3D Anomaly Detection (NeurIPS 2025 · 0 citations) |
| Vary the number of local sets | KITTI · RR (%) | 98.02 to 98.20 | No clear effect | A robust inlier identification algorithm for point cloud registration via $\mathbf{\ell_0}$-minimization (NeurIPS 2024 · 0 citations) |
What reviewers will expect
Datasets and baselines named in the 5 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- 3DMatch · 3 of 5 papers
- 3DLoMatch · 2 of 5 papers
Baselines
- TEASER++ · 2 of 5 papers
- SC2-PCR · 2 of 5 papers
- MAC · 2 of 5 papers
- PointDSC · 2 of 5 papers
- VBReg · 2 of 5 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 16 papers; from OpenAlex and the papers' affiliations).
- Tsinghua University · 3 papers
- National University of Defense Technology · 2 papers
- Sun Yat-sen University · 2 papers
- Wuhan University · 2 papers
- Huazhong University of Science and Technology · 2 papers
Use it
Download BibTeX of 28 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Learned feature-based registration (66 papers)
Elsewhere in the topic
- Non-rigid point set registration (42 papers)
- Quantum and certifiable correspondence (8 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).