Research map / Point clouds / Registration and matching
FINet: Dual Branches Feature Interaction for Partial-to-Partial Point Cloud Registration
Hao Xu, Nianjin Ye, Guanghui Liu, Bing Zeng, Shuaicheng 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 |
|---|---|---|---|
| add multi-level feature interactions (PFI and GFI) (ablation) | FINet · ModelNet40 unseen shapes and categories | RMSE(R): 4.857 -> 4.023 | Improved |
| add PFI to the single-branch setup (ablation) | FINet · ModelNet40 ablation | RMSE(R): 4.921 -> 4.368 | Improved |
| add GFI to the single-branch setup (ablation) | FINet · ModelNet40 ablation | RMSE(R): 4.921 -> 4.792 | Improved |
| add PFI to the dual-branch setup (ablation) | FINet · ModelNet40 ablation | RMSE(R): 4.857 -> 4.308 | Improved |
| add GFI to the dual-branch setup (ablation) | FINet · ModelNet40 ablation | RMSE(R): 4.857 -> 4.220 | Improved |
| combine PFI and GFI rather than using PFI alone (ablation) | FINet dual branches · ModelNet40 ablation | RMSE(R): 4.308 -> 4.023 | Improved |
| add point-wise feature dropout loss (ablation) | FINet · ModelNet40 ablation | RMSE(R): 4.023 -> 3.902 | Improved |
| add transformation sensitivity loss (ablation) | FINet · ModelNet40 ablation | RMSE(R): 3.902 -> 3.841 | Improved |
| use dual branches rather than single branch without additional supervising (ablation) | FINet · ModelNet40 ablation | RMSE(R): 4.921 -> 4.857 | Improved |
| regress saliency points instead of omitting the explicit SP objective (ablation) | FINet dual branches · ModelNet40 ablation | RMSE(R): 4.023 -> 3.955 | Improved |
| add saliency-point regression in the single-branch setup (ablation) | FINet single branch · ModelNet40 ablation | RMSE(R): 4.427 -> 4.481 | Hurt |
| use FINet rather than OMNet | ModelNet40 unseen shapes · ModelNet40 unseen shapes, OS and TS | RMSE(R): 0.771 -> 0.694; 1.384 -> 1.267 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- OMNet — OMNet: Learning Overlapping Mask for Partial-to-Partial Point Cloud Registration (ICCV 2021) · better on 6 of 7 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
- Dual Focus-Attention Transformer for Robust Point Cloud Registration (CVPR 2025 · similarity 0.88)
- Dynamic Cues-Assisted Transformer for Robust Point Cloud Registration (CVPR 2024 · similarity 0.85)
- PSReg: Prior-guided Sparse Mixture of Experts for Point Cloud Registration (AAAI 2025 · similarity 0.85)
- FreeReg: Image-to-Point Cloud Registration Leveraging Pretrained Diffusion Models and Monocular Depth Estimators (ICLR 2024 · similarity 0.85)
- PointMC: Multi-instance Point Cloud Registration based on Maximal Cliques (ICML 2024 · similarity 0.85)
- Partial Point Cloud Registration with Multi-view 2D Image Learning (AAAI 2025 · similarity 0.85)
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.
