Research map / 3D vision / Depth and stereo
Global Structure-from-Motion Revisited
Linfei Pan, Daniel Barath, Marc Pollefeys, Johannes L Schönberger
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 |
|---|---|---|---|
| replace relative translation constraints with point-only global positioning (ablation) | GLOMAP · ETH3D MVS (DSLR) | AUC @ 1deg: 73.5 -> 80.8 | Improved |
| use point-only global positioning instead of points plus relative translations (ablation) | GLOMAP · IMC 2023 | AUC @ 3deg: 68.6 -> 69.6 | Improved |
| replace Theia LUD translation averaging with joint camera-and-point global positioning (ablation) | GLOMAP · IMC 2023 | AUC @ 3deg: 64.3 -> 69.6 | Improved |
| use points-only positioning with faster runtime than LUD (ablation) | GLOMAP · IMC 2023 | Time (s): 987.2 -> 497.3 | Improved |
| use GLOMAP instead of COLMAP on ETH3D SLAM | global SfM · ETH3D SLAM (average) | Time (s): 1115.4 -> 133.5 | Improved |
| use GLOMAP instead of COLMAP on LaMAR | global SfM · LaMAR (average) | Recall @ 1m: 32.0 -> 49.1 | Improved |
| use GLOMAP instead of COLMAP on IMC 2023 | global SfM · IMC 2023 (average) | AUC @ 3deg: 65.3 -> 69.6 | Improved |
| use GLOMAP instead of COLMAP on MIP360 | global SfM · MIP360 (average) | AUC @ 3deg: 96.5 -> 97.5 | No clear effect |
| use GLOMAP instead of COLMAP on ETH3D MVS DSLR | global SfM · ETH3D MVS DSLR (average) | AUC @ 1deg: 79.2 -> 80.8 | No clear effect |
| run GLOMAP on rotationally symmetric scene exhibition_hall | GLOMAP · ETH3D MVS DSLR (exhibition_hall) | AUC @ 1deg: 25.5 vs COLMAP 85.4 | Hurt |
Closest papers by meaning
- VGGSfM: Visual Geometry Grounded Deep Structure From Motion (CVPR 2024 · similarity 0.89)
- Light3R-SfM: Towards Feed-forward Structure-from-Motion (CVPR 2025 · similarity 0.87)
- MP-SfM: Monocular Surface Priors for Robust Structure-from-Motion (CVPR 2025 · similarity 0.86)
- Detector-Free Structure from Motion (CVPR 2024 · similarity 0.86)
- Structure-from-Motion with a Non-Parametric Camera Model (CVPR 2025 · similarity 0.85)
- Leveraging Camera Triplets for Efficient and Accurate Structure-from-Motion (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.
