Research map / Point clouds / Registration and matching
A Performance Evaluation of Correspondence Grouping Methods for 3D Rigid Data Matching
Jiaqi Yang, Ke Xian, Peng Wang, Yanning Zhang
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 |
|---|---|---|---|
| use GTM instead of other grouping methods under clutter and occlusion | correspondence grouping · U3OR | precision | Improved |
| use GTM at high overlap | correspondence grouping · U3M | precision: approximately 30% and 20% with 90% and 80% overlap | Improved |
| use ST on clutter and occlusion data rather than noisy full shapes | ST · U3OR vs B3R | precision/recall | Hurt |
| increase Gaussian noise beyond 0.2 pr | SI · B3R noise | precision/recall | Hurt |
| decrease point density (down-sampling) for SHOT-based NNSR and SI | NNSR, SI · B3R density variation | recall | Hurt |
| increase number of initial matches above 1000 | SI, CV · U3OR | precision: low below 1000, improves quickly above 1000 | Improved |
| use voting-based CV and SI for recall | correspondence grouping · U3M | recall | Improved |
| use GC instead of RANSAC for registration | rigid registration · U3OR clutter and occlusion | registration success | Improved |
| use RCS descriptor instead of SHOT or LFSH | all methods · U3OR | precision/recall | Improved |
| use ISS+LFSH combination | most methods · U3M | precision/recall | Improved |
| use SS (descriptor similarity only) for registration | SS · U3OR | registration success | Hurt |
| raise inlier distance threshold epsilon | all methods · U3OR | precision | Improved |
Closest papers by meaning
- Partial-to-Partial Shape Matching with Geometric Consistency (CVPR 2024 · similarity 0.87)
- DV-Matcher: Deformation-based Non-rigid Point Cloud Matching Guided by Pre-trained Visual Features (CVPR 2025 · similarity 0.87)
- PointMC: Multi-instance Point Cloud Registration based on Maximal Cliques (ICML 2024 · similarity 0.87)
- Structure-Aware Correspondence Learning for Relative Pose Estimation (CVPR 2025 · similarity 0.87)
- EchoMatch: Partial-to-Partial Shape Matching via Correspondence Reflection (CVPR 2025 · similarity 0.87)
- Scalable 3D Registration via Truncated Entry-wise Absolute Residuals (CVPR 2024 · 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.
