Research map / Point clouds / Registration and matching
Match Normalization: Learning-Based Point Cloud Registration for 6D Object Pose Estimation in the Real World
Zheng Dang, Lizhou Wang, Yu Guo, Mathieu Salzmann
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 Match Normalization (ablation) | Ours-DCP · TUD-L | Rotation mAP 5 deg: 0.21 -> 0.91 | Improved |
| add Match Normalization (ablation) | Ours-DCP · TUD-L | BOP AR: 0.27 -> 0.90 | Improved |
| add Match Normalization (ablation) | Ours-DCP · TUD-L | predicted matches: 24.84 -> 276.10 | Improved |
| add Match Normalization (ablation) | Ours-DCP · TUD-L | true inlier matches: 10.44 -> 262.96 | Improved |
| replace SVD loss with NLL loss (ablation) | DCP(v2) · ModelNet40 (full rotation range) | Rotation mAP 5 deg: 0.00 -> 0.35 | Improved |
| replace SVD loss with NLL loss (ablation) | DCP(v2) · ModelNet40 (45 deg range) | Rotation mAP 5 deg: 0.36 -> 0.72 | Improved |
| add ICP post-processing | Ours-DCP · TUD-L | BOP AR: 0.801 -> 0.903 | Improved |
| add ICP post-processing | Ours-DCP · LINEMOD | BOP AR: 0.446 -> 0.678 | Improved |
| add ICP post-processing | Ours-DCP · Occluded-LINEMOD | BOP AR: 0.365 -> 0.544 | Improved |
| add Match Normalization and NLL loss (Ours-DCP) | DCP(v2) · TUD-L | BOP AR: 0.038 -> 0.801 | Improved |
| add Match Normalization | IDAM · TUD-L | BOP AR: 0.091 -> 0.444 | Improved |
| add Match Normalization | IDAM · LINEMOD | BOP AR: 0.067 -> 0.184 | Improved |
Closest papers by meaning
- Learning-Based Point Cloud Registration for 6D Object Pose Estimation in the Real World (ECCV 2022 · similarity 0.98)
- Register Any Point: Scaling 3D Point Cloud Registration by Flow Matching (ECCV 2026 · similarity 0.90)
- Learning Symmetry-Aware Geometry Correspondences for 6D Object Pose Estimation (ICCV 2023 · similarity 0.90)
- PointNetLK Revisited (CVPR 2021 · similarity 0.90)
- DV-Matcher: Deformation-based Non-rigid Point Cloud Matching Guided by Pre-trained Visual Features (CVPR 2025 · similarity 0.88)
- Zero-shot RGB-D Point Cloud Registration with Pre-trained Large Vision Model (CVPR 2025 · 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.
