Research map / 3D vision / Depth and stereo
Uncertainty Based Camera Model Selection
Michal Polic, Stanislav Steidl, Cenek Albl, Zuzana Kukelova, Tomas Pajdla
Check 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 LACS across reprojection thresholds instead of selecting by one ACS threshold (ablation) | ACS camera model selection · Synthetic camera model selection, K=15 | success ratio: 0.76 -> 0.93 | Improved |
| Use LACS across reprojection thresholds instead of ACS at δ=2.0px (ablation) | ACS camera model selection · Synthetic camera model selection, K=10 | success ratio: 0.68 -> 0.83 | Improved |
| Use LACS across reprojection thresholds instead of ACS at δ=2.0px (ablation) | ACS camera model selection · Synthetic camera model selection, K=5 | success ratio: 0.48 -> 0.68 | Improved |
| Account for observation uncertainty when classifying inliers | SfM observation inlier filtering · Synthetic data with randomly permuted correspondences, δ=1.6px | true positive inliers: increase of about 19.4% | Improved |
| Account for observation uncertainty when classifying inliers | SfM observation inlier filtering · Synthetic data with systematically permuted correspondences, δ=1.6px | true positive inliers: increased by 31.7% | Improved |
| Use a fixed reprojection-error threshold without observation uncertainty (ablation) | SfM observation inlier filtering · Synthetic data with systematically permuted correspondences, δ=4px | false positive inliers: increases the false positives by about 9.7% | Hurt |
| Use AC-based automatic camera model selection rather than an unsuitable manually chosen model | COLMAP SfM · terrains rig and ETH datasets | reconstruction quality | Improved |
| Use AC-based automatic camera model selection rather than an unsuitable camera model | COLMAP SfM · terrains rig and ETH datasets | SfM run-time | Improved |
Closest papers by meaning
- Fast and Accurate Camera Covariance Computation for Large 3D Reconstruction (ECCV 2018 · similarity 0.90)
- View-graph Selection Framework for SfM (ECCV 2018 · similarity 0.88)
- MP-SfM: Monocular Surface Priors for Robust Structure-from-Motion (CVPR 2025 · similarity 0.86)
- Leveraging Camera Triplets for Efficient and Accurate Structure-from-Motion (CVPR 2024 · similarity 0.85)
- Structure-from-Motion with a Non-Parametric Camera Model (CVPR 2025 · similarity 0.85)
- Variational Multi-scale Representation for Estimating Uncertainty in 3D Gaussian Splatting (NeurIPS 2024 · similarity 0.84)
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.
