Research map / Point clouds / Registration and matching
MCI-Net: A Robust Multi-Domain Context Integration Network for Point Cloud Registration
Shuyuan Lin, Wenwu Peng, Junjie Huang, Qiang Qi, Miaohui Wang, Jian Weng
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 graph neighborhood aggregation module (GNAM) (ablation) | MCI-Net · 3DMatch | Registration Recall: 95.2 -> 96.4 | Improved |
| add GNAM (ablation) | MCI-Net · 3DLoMatch | Registration Recall: 78.9 -> 79.7 | Improved |
| add intra-domain feature decoupling (IFD) (ablation) | MCI-Net · 3DMatch | Registration Recall: 95.3 -> 96.4 | Improved |
| add intra-domain feature decoupling (IFD) (ablation) | MCI-Net · 3DLoMatch | Registration Recall: 79.3 -> 79.7 | Improved |
| add inter-domain cross-attention fusion (ICI) (ablation) | MCI-Net · 3DMatch | Registration Recall: 95.7 -> 96.4 | Improved |
| add inter-domain cross-attention fusion (ICI) (ablation) | MCI-Net · 3DLoMatch | Registration Recall: 79.5 -> 79.7 | Improved |
| use DISM instead of RANSAC (ablation) | pose estimator · 3DMatch | Registration Recall: 94.6 -> 96.4 | Improved |
| use DISM instead of LGR (ablation) | pose estimator · 3DMatch | Registration Recall: 94.9 -> 96.4 | Improved |
| use DISM instead of LGR (ablation) | pose estimator · 3DLoMatch | Registration Recall: 79.1 -> 79.8 | Improved |
| use MCI-Net features instead of PSReg | PSReg · 3DLoMatch | Feature Matching Recall: 86.6 vs 85.1 (2500 samples, Table 1) | Hurt |
| use MCI-Net on KITTI instead of GeoTrans | GeoTrans · KITTI Odometry | Relative Translation Error: 6.2 -> 5.0 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 on most shared results.
- PSReg — PSReg: Prior-guided Sparse Mixture of Experts for Point Cloud Registration (AAAI 2025) · better on 2 of 3 shared results
- PEAL — PEAL: Prior-Embedded Explicit Attention Learning for Low-Overlap Point Cloud Registration (CVPR 2023) · better on 2 of 3 shared results
- Cross-PCR — Cross-PCR: A Robust Cross-Source Point Cloud Registration Framework (AAAI 2025) · better on 2 of 2 shared results
- BUFFER — BUFFER: Balancing Accuracy, Efficiency, and Generalizability in Point Cloud Registration (CVPR 2023) · better on 1 of 1 shared result
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
- PointDSC: Robust Point Cloud Registration Using Deep Spatial Consistency (CVPR 2021 · similarity 0.89)
- Feature Interactive Representation for Point Cloud Registration (ICCV 2021 · similarity 0.89)
- Mining and Transferring Feature-Geometry Coherence for Unsupervised Point Cloud Registration (NeurIPS 2024 · similarity 0.88)
- HRegNet: A Hierarchical Network for Large-Scale Outdoor LiDAR Point Cloud Registration (ICCV 2021 · similarity 0.88)
- MHopReg: Efficient Hierarchical Multi-Hop Graph Search for Point Cloud Registration (CVPR 2026 · similarity 0.88)
- Robust Point Cloud Registration Framework Based on Deep Graph Matching (CVPR 2021 · similarity 0.88)
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.
