Research map / Point clouds / Registration and matching
Feature Interactive Representation for Point Cloud Registration
Bingli Wu, Jie Ma, Gaojie Chen, Pei An
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 Combined Feature Encoder to baseline (ablation) | BS · ModelNet40 noisy unseen objects, MOR 82% (ablation) | RMSE(R): 14.30 -> 12.67 | Improved |
| add Local Interaction Unit to BS + CFE (ablation) | BS + CFE · ModelNet40 noisy unseen objects, MOR 82% (ablation) | RMSE(R): 12.67 -> 11.48 | Improved |
| add Global Interaction Unit to BS + CFE (ablation) | BS + CFE · ModelNet40 noisy unseen objects, MOR 82% (ablation) | RMSE(R): 12.67 -> 11.74 | Improved |
| add GIU on top of CFE and LIU (ablation) | BS + CFE + LIU · ModelNet40 noisy unseen objects, MOR 82% (ablation) | RMSE(R): 11.48 -> 5.76 | Improved |
| add LIU on top of CFE and GIU (ablation) | BS + CFE + GIU · ModelNet40 noisy unseen objects, MOR 82% (ablation) | RMSE(R): 11.74 -> 5.76 | Improved |
| add LIU and GIU on top of CFE (ablation) | BS + CFE · ModelNet40 noisy unseen objects, MOR 82% (ablation) | MAE(t): 0.071 -> 0.013 | Improved |
| use type 2 local graph instead of type 1 in LIU (ablation) | LIU · ModelNet40 noisy unseen objects, MOR 82% (ablation) | RMSE(R): 10.03 -> 5.76 | Improved |
| use association form 3 (node feature and difference) instead of form 1 (ablation) | LIU · ModelNet40 noisy unseen objects, MOR 82% (ablation) | RMSE(R): 7.30 -> 5.76 | Improved |
| use association form 3 instead of form 2 (ablation) | LIU · ModelNet40 noisy unseen objects, MOR 82% (ablation) | RMSE(R): 6.57 -> 5.76 | Improved |
| combine maxpool and meanpool global features instead of maxpool only (ablation) | GIU · ModelNet40 noisy unseen objects, MOR 82% (ablation) | RMSE(R): 6.69 -> 5.76 | Improved |
| combine maxpool and meanpool global features instead of meanpool only (ablation) | GIU · ModelNet40 noisy unseen objects, MOR 82% (ablation) | RMSE(R): 6.04 -> 5.76 | Improved |
| lower overlap rate from 96% to 82% | FIRE-Net · ModelNet40 unseen objects | RMSE(R) (deg): 0.95 -> 2.21 | Hurt |
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.
- PRNet — PRNet: Self-Supervised Learning for Partial-to-Partial Registration (NeurIPS 2019) · better on 2 of 2 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
- Mining and Transferring Feature-Geometry Coherence for Unsupervised Point Cloud Registration (NeurIPS 2024 · similarity 0.88)
- PointMC: Multi-instance Point Cloud Registration based on Maximal Cliques (ICML 2024 · similarity 0.87)
- Dual Focus-Attention Transformer for Robust Point Cloud Registration (CVPR 2025 · similarity 0.87)
- FreeReg: Image-to-Point Cloud Registration Leveraging Pretrained Diffusion Models and Monocular Depth Estimators (ICLR 2024 · similarity 0.87)
- PSReg: Prior-guided Sparse Mixture of Experts for Point Cloud Registration (AAAI 2025 · similarity 0.87)
- PointCFormer: A Relation-Based Progressive Feature Extraction Network for Point Cloud Completion (AAAI 2025 · similarity 0.86)
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.
