Research map / Point clouds / Registration and matching
Robust Point Cloud Based Reconstruction of Large-Scale Outdoor Scenes
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 Cauchy-Uniform EM with feature matches as constraints instead of Choi et al.'s covariance-based method | Robust point-cloud reconstruction · Oxford Robotcar 1km | mean pose distance (m): 1.97 -> 1.34 | Helped |
| Use Cauchy-Uniform EM with feature matches as constraints instead of Choi et al.'s covariance-based method | Robust point-cloud reconstruction · Oxford Robotcar city-scale | mean pose distance (m): 50.92 -> 2.45 | Helped |
| Encode feature-match uncertainty in a covariance matrix rather than use identity covariance | Choi et al. baseline · Oxford Robotcar 1km | mean pose distance (m): 123.24 -> 1.97 | Helped |
| Encode feature-match uncertainty in a covariance matrix rather than use identity covariance | Choi et al. baseline · Oxford Robotcar city-scale | mean pose distance (m): 207.93 -> 50.92 | Helped |
| Use Gaussian-Uniform EM instead of Choi et al.'s robust optimization | Indoor reconstruction · Augmented ICL-NUIM average | mean surface distance (m): 0.05 -> 0.06 | Hurt |
| Use Gaussian-Uniform EM instead of Choi et al.'s pruning method | Indoor loop-closure pruning · Augmented ICL-NUIM average | loop-closure recall (%): 57.8 -> 58.1 | No clear effect |
| Use Gaussian-Uniform EM instead of Choi et al.'s pruning method | Indoor loop-closure pruning · Augmented ICL-NUIM average | loop-closure precision (%): 97.7 -> 95.4 | Hurt |
Closest papers by meaning
- A robust inlier identification algorithm for point cloud registration via $\mathbf{\ell_0}$-minimization (NeurIPS 2024 · similarity 0.88)
- PointCVaR: Risk-Optimized Outlier Removal for Robust 3D Point Cloud Classification (AAAI 2024 · similarity 0.86)
- Inlier Confidence Calibration for Point Cloud Registration (CVPR 2024 · similarity 0.86)
- Hyperbolic-Constraint Point Cloud Reconstruction from Single RGB-D Images (AAAI 2025 · similarity 0.85)
- APML: Adaptive Probabilistic Matching Loss for Robust 3D Point Cloud Reconstruction (NeurIPS 2025 · similarity 0.85)
- Scalable 3D Registration via Truncated Entry-wise Absolute Residuals (CVPR 2024 · similarity 0.85)
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.