Research map / Point clouds / Registration and matching
Guaranteed Outlier Removal for Point Cloud Registration with Correspondences
Álvaro Parra, Tat-Jun Chin
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 GORE preprocessing before BnB | BnB · 6 DoF point cloud registration (real) | runtime: about half of the time required by BnB alone | Improved |
| add GORE preprocessing before RANSAC | RANSAC · 6 DoF point cloud registration (real) | runtime: two orders of magnitude smaller | Improved |
| add GORE preprocessing before RANSAC | RANSAC · synthetic rotation search, low outlier rate | runtime: RANSAC faster than GORE at low outlier rates | Hurt |
| add GORE preprocessing before RANSAC | RANSAC · synthetic rotation search, outlier rate >= 90% | runtime: significant gain for >=90% outliers | Improved |
| warm-start BnB with RANSAC lower bound (RGORE+BnB) (ablation) | GORE+BnB · synthetic rotation search | runtime: only marginally reduced the total runtime at low outlier rates | No clear effect |
| add BnB step inside GORE (6 DoF) (ablation) | GORE · Remote sensing real data (N=500) | removed outliers: 43 -> 353 | Improved |
| add BnB step inside GORE (6 DoF) (ablation) | GORE · Mian real data (N=500) | removed outliers: 422 -> 435 | Improved |
| add BnB step inside GORE (6 DoF) (ablation) | GORE · Mining real data (N=2000) | runtime (s): 1.284 -> 11.034 | Hurt |
| add BnB step inside GORE (6 DoF) (ablation) | GORE · synthetic registration, high outlier ratios | runtime and pruning: extra cost of one order of magnitude; slightly improved pruning | Mixed |
| use GORE rotation estimate instead of RANSAC | RANSAC · synthetic rotation search | angular error: error <= 0.05 degrees | Improved |
| add GORE preprocessing before SVD least squares | SVD rotation estimation · synthetic rotation search | angular error: non-negligible error remains after GORE | No clear effect |
| use GORE instead of Albarelli's method for outlier reduction | Albarelli · mining and remote sensing real data | precision and recall: recall always one; Albarelli precision and recall were 0 | Improved |
Closest papers by meaning
- A New Outlier Removal Strategy Based on Reliability of Correspondence Graph for Fast Point Cloud Registration (TPAMI 2022 · similarity 0.95)
- A Practical O(N2) Outlier Removal Method for Point Cloud Registration (TPAMI 2021 · similarity 0.95)
- QGORE: Quadratic-Time Guaranteed Outlier Removal for Point Cloud Registration (TPAMI 2023 · similarity 0.93)
- A robust inlier identification algorithm for point cloud registration via $\mathbf{\ell_0}$-minimization (NeurIPS 2024 · similarity 0.93)
- Inlier Confidence Calibration for Point Cloud Registration (CVPR 2024 · similarity 0.91)
- Scalable 3D Registration via Truncated Entry-wise Absolute Residuals (CVPR 2024 · similarity 0.89)
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.
