Research map / Point clouds / Registration and matching
Non-rigid point set registration
Align point sets with rigid and non-rigid transformations using coherent drift, graph consistency and transformers.
42 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Registration and matching, keyword cluster “rigid · point registration · non”
State History Tried Checklist Preprints Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
9 papers in 2023–24 and 6 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 25.7% to 13.0%.
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
- 2016 · Gravitational Approach for Point Set Registration (CVPR 2016 · 46 citations)
- 2017 · Convex Global 3D Registration With Lagrangian Duality (CVPR 2017 · 109 citations)
- 2018 · Density Adaptive Point Set Registration (CVPR 2018 · 61 citations)
- 2019 · FilterReg: Robust and Efficient Probabilistic Point-Set Registration Using Gaussian Filter and Twist Parameterization (CVPR 2019 · 133 citations)
- 2020 · RPM-Net: Robust Point Matching Using Learned Features (CVPR 2020 · 623 citations)
- 2021 · Locally Aware Piecewise Transformation Fields for 3D Human Mesh Registration (CVPR 2021 · 66 citations)
- 2022 · Lepard: Learning Partial Point Cloud Matching in Rigid and Deformable Scenes (CVPR 2022 · 183 citations)
- 2023 · Rotation-Invariant Transformer for Point Cloud Matching (CVPR 2023 · 143 citations)
- 2024 · Scalable 3D Registration via Truncated Entry-wise Absolute Residuals (CVPR 2024 · 12 citations)
- 2025 · GRICP: Granular-Ball Iterative Closest Point with Multikernel Correntropy for Point Cloud Fine Registration (AAAI 2025 · 4 citations)
- 2026 · Generalized-CVO: Fast and Correspondence-Free Local Point Cloud Registration with Second Order Riemannian Optimization (CVPR 2026 · 0 citations)
Earliest papers here
- Gravitational Approach for Point Set Registration (CVPR 2016 · 46 citations)
- A Probabilistic Framework for Color-Based Point Set Registration (CVPR 2016 · 43 citations)
- Context-Aware Gaussian Fields for Non-Rigid Point Set Registration (CVPR 2016 · 29 citations)
What has been tried
39 papers here report 356 measured changes (a component added, removed or swapped, compared with a baseline): helped 69%, no clear effect 4%, hurt 21%, mixed 6%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Use TEAR-1 rather than CM-1 as the branch-and-bound objective | Synthetic data, 95% outliers · F1-score | Helped | Scalable 3D Registration via Truncated Entry-wise Absolute Residuals (CVPR 2024 · 12 citations) | |
| Reintegrate predicted overlap scores into attention learning | 3DLoMatch · RR (%) | 0.791 -> 0.812 | Helped | Multiway Point Cloud Mosaicking with Diffusion and Global Optimization (CVPR 2024 · 11 citations) |
| use Coupled Laplacian eigenmaps instead of Euclidean point distances | MVTec 3D-AD · area under PRO curve (limit 0.3), mean | 0.703 -> 0.748 | Helped | Coupled Laplacian Eigenmaps for Locally-Aware 3D Rigid Point Cloud Matching (CVPR 2024 · 6 citations) |
| Represent rotation with quaternions instead of Euler angles | ModelNet40, completely overlapping · MAE(R) | 13.7668 -> 1.6935 | Helped | Where Precision Meets Efficiency: Transformation Diffusion Model for Point Cloud Registration (AAAI 2025 · 4 citations) |
| Add pose refinement (+R) | PhotoTourism · AUC@10 | 73.73 -> 77.00 | Helped | Practical Solutions to the Relative Pose of Three Calibrated Cameras (CVPR 2025 · 0 citations) |
What reviewers will expect
Datasets and baselines named in the 7 papers of this approach from 2024–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- 3DMatch · 4 of 7 papers
- KITTI · 3 of 7 papers
Baselines
- FGR · 3 of 7 papers
- ICP · 2 of 7 papers
- REGTR · 2 of 7 papers
- FPFH · 2 of 7 papers
Recent preprints
Loading recent arXiv preprints…
Use it
Download BibTeX of 29 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
- Quantum and certifiable correspondence (8 papers)
Elsewhere in the topic
- Learned feature-based registration (66 papers)
- Robust inlier selection for registration (29 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).