Research map / 3D vision / Human pose estimation
Deep Global Registration
Check 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 Deep Global Registration instead of RANSAC-8M | 3DMatch pairwise registration · 3DMatch test | recall: 74.9% -> 91.3% | Helped |
| use Deep Global Registration instead of RANSAC-8M | 3DMatch pairwise registration · 3DMatch test | TE (cm): 8.96 -> 7.34 | Helped |
| use Deep Global Registration instead of RANSAC-8M | 3DMatch pairwise registration · 3DMatch test | RE (deg): 2.92 -> 2.43 | Helped |
| remove safeguard registration (ablation) | Deep Global Registration · 3DMatch test | recall: 85.2% -> 91.3% | Hurt |
| remove safeguard registration (ablation) | Deep Global Registration · 3DMatch test | TE (cm): 7.34 -> 7.73 | Hurt |
| remove safeguard registration (ablation) | Deep Global Registration · 3DMatch test | RE (deg): 2.43 -> 2.58 | Hurt |
| add ICP fine-grained pose adjustment (ablation) | Deep Global Registration · KITTI test split | TE (cm): 21.7 -> 3.46 | Helped |
| add ICP fine-grained pose adjustment (ablation) | Deep Global Registration · KITTI test split | recall: 96.9% -> 98.0% | Helped |
| add ICP fine-grained pose adjustment (ablation) | Deep Global Registration · KITTI test split | RE (deg): 0.34 -> 0.14 | Helped |
| use multi-way registration with the proposed pairwise method rather than the tested alternatives | Augmented ICL-NUIM reconstruction · Augmented ICL-NUIM | Avg. Rank: 2.5 -> 2 | Helped |
Closest papers by meaning
- 3DRegNet: A Deep Neural Network for 3D Point Registration (CVPR 2020 · similarity 0.86)
- 3D Local Features for Direct Pairwise Registration (CVPR 2019 · similarity 0.86)
- CARL: A Framework for Equivariant Image Registration (CVPR 2025 · similarity 0.86)
- FUSER: Feed-Forward Multiview 3D Registration Transformer and SE(3)$^N$ Diffusion Refinement (CVPR 2026 · similarity 0.85)
- Fine-To-Coarse Global Registration of RGB-D Scans (CVPR 2017 · similarity 0.84)
- Scalable 3D Registration via Truncated Entry-wise Absolute Residuals (CVPR 2024 · similarity 0.83)
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.