Research map / Federated learning and privacy / Privacy attacks and secure learning
Privacy Preserving Structure-from-Motion
Marcel Geppert, Viktor Larsson, Pablo Speciale, Johannes L. Schönberger, Marc Pollefeys
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 privacy-preserving random line features instead of point features | COLMAP · Internet photos: Madrid Metropolis | registered images: 279 -> 377 | Improved |
| use privacy-preserving random line features instead of point features | COLMAP · Internet photos: Gendarmenmarkt | registered images: 825 -> 810 | No clear effect |
| use privacy-preserving random line features instead of point features | COLMAP · Internet photos: Madrid Metropolis | median point reprojection error: 0.45 -> 1.13 | Hurt |
| use privacy-preserving random line features instead of point features | COLMAP · Internet photos: Tower of London | median point reprojection error: 0.38 -> 0.54 | Hurt |
| use privacy-preserving random line features instead of point features | COLMAP · Internet photos: Alamo | mean track length: 12.8 -> 21.9 | Improved |
| use privacy-preserving random line features instead of point features | COLMAP · Internet photos: Tower of London | number of 3D points: 146k -> 93k | Hurt |
| use privacy-preserving lines instead of points | COLMAP · Strecha castle-P30 | position error (cm): 3cm -> 4.00 | Hurt |
| use privacy-preserving lines instead of points | COLMAP · Speciale et al. six datasets | recall at 25cm threshold: 90 | Mixed |
| select 50% of features as gravity-aligned instead of a smaller share (ablation) | privacy preserving SfM initialization · Strecha and Speciale datasets | initialization tracks | Improved |
Closest papers by meaning
- Privacy Preserving Image-Based Localization (CVPR 2019 · similarity 0.93)
- Privacy Preserving Localization via Coordinate Permutations (ICCV 2023 · similarity 0.92)
- Privacy Preserving Visual SLAM (ECCV 2020 · similarity 0.92)
- Efficient Privacy-Preserving Visual Localization Using 3D Ray Clouds (CVPR 2024 · similarity 0.89)
- Gaussian Splatting Feature Fields for (Privacy-Preserving) Visual Localization (CVPR 2025 · similarity 0.87)
- GPVK-VL: Geometry-Preserving Virtual Keyframes for Visual Localization under Large Viewpoint Changes (CVPR 2025 · 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.
