Research map / Point clouds / Registration and matching
Unsupervised Deep Probabilistic Approach for Partial Point Cloud Registration
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use UDPReg instead of SGP | 3DMatch registration · 3DMatch | RR: 85.5% -> 91.4% | Helped |
| Use UDPReg instead of SGP | 3DLoMatch registration · 3DLoMatch | RR: 39.4% -> 64.3% | Helped |
| Combine self-consistency, cross-consistency, and local contrastive losses (ablation) | UDPReg · ModelNet | RRE: 3.421 -> 1.331 | Helped |
| Add local contrastive loss to CC + SC (ablation) | UDPReg · ModelLoNet | RRE: 5.229 -> 3.578 | Helped |
| Use SC only instead of the full loss combination (ablation) | UDPReg · ModelNet | RRE: 1.331 -> 5.898 | Hurt |
| Use LC only instead of the full loss combination (ablation) | UDPReg · ModelLoNet | RRE: 3.578 -> 8.790 | Hurt |
| Use cluster-to-point hierarchical correspondence matching instead of cluster-only matching (ablation) | UDPReg · ModelNet | RRE: 3.932 -> 1.331 | Helped |
| Use cluster-to-point hierarchical correspondence matching instead of direct point matching (ablation) | UDPReg · ModelLoNet | RRE: 4.264 -> 3.578 | Helped |
| Vary cluster count from 16 to 96 (ablation) | UDPReg · ModelNet and ModelLoNet | RRE | No clear effect |
| Set cluster count to 64 (ablation) | UDPReg · ModelLoNet | RRE: 3.598 -> 3.578 | Helped |
Closest papers by meaning
- PSReg: Prior-guided Sparse Mixture of Experts for Point Cloud Registration (AAAI 2025 · similarity 0.89)
- FreeReg: Image-to-Point Cloud Registration Leveraging Pretrained Diffusion Models and Monocular Depth Estimators (ICLR 2024 · similarity 0.88)
- 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 0.88)
- Cross-PCR: A Robust Cross-Source Point Cloud Registration Framework (AAAI 2025 · similarity 0.87)
- HybridReg: Robust 3D Point Cloud Registration with Hybrid Motions (AAAI 2025 · similarity 0.87)
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.