Research map / 3D vision / Depth and stereo
Privacy-Preserving Representations Are Not Enough: Recovering Scene Content From Camera Poses
Read the paperCheck 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 |
|---|---|---|---|
| Query the localization service with multi-view object images and classify presence using pose-alignment inlier-ratio statistics | HLoc visual localization · IKEA-Scenes | object presence precision and recall: approx. 0.4/0.6; random guessing precision 0.19 | Helped |
| Use Superpoint features with SuperGlue matching instead of less robust feature and matcher combinations | HLoc visual localization · IKEA-Scenes object presence classification | precision and recall | Mixed |
| Reject queries when too few distinct scene objects contribute inliers | Localization-service defense · RIO10 | retention of genuine queries while rejecting malicious queries | Mixed |
| Use stricter RANSAC inlier thresholds | HLoc localization server · RIO10 object alignment examples | qualitative object placement | No clear effect |
Closest papers by meaning
- Privacy Preserving Image-Based Localization (CVPR 2019 · similarity 0.92)
- Efficient Privacy-Preserving Visual Localization Using 3D Ray Clouds (CVPR 2024 · similarity 0.90)
- 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 0.82)
- SecureGS: Boosting the Security and Fidelity of 3D Gaussian Splatting Steganography (ICLR 2025 · similarity 0.82)
- GUIGuard-Bench: Toward a General Evaluation for Privacy-Preserving GUI Agents (NeurIPS 2026 · similarity 0.82)
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.