Research map / Federated learning and privacy / Privacy attacks and secure learning
VPR-Cloak: A First Look at Privacy Cloak Against Visual Place Recognition
Shuting Dong, Mingzhi Chen, Feng Lu, Hao Yu, Guanghao Li, Zhe Wu, Ming Tang, Chun Yuan
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 |
|---|---|---|---|
| add saliency-aware prior to perturbation optimization (ablation) | VPR-Cloak · Pitts30k | LPIPS: 0.183 -> 0.057 | Improved |
| add saliency-aware prior (ablation) | VPR-Cloak · Pitts30k | Avg Epsilon: 5.69 -> 2.84 | Improved |
| add frequency-domain optimization to spatial optimization (ablation) | VPR-Cloak · Pitts30k (Crica) | Delta R@1: 51.6 -> 57.1 | Improved |
| use frequency-domain only instead of spatial only (ablation) | VPR-Cloak · Pitts30k (Crica) | Delta R@1: 51.6 -> 27.5 | Hurt |
| add frequency optimization to spatial (visual quality) (ablation) | VPR-Cloak · Pitts30k | PSNR: 32.43 -> 32.42 | Hurt |
| use frequency-only perturbation for image quality (ablation) | VPR-Cloak · Pitts30k | PSNR: 32.43 -> 33.83 | Improved |
| use VPR-Cloak instead of MultiANDA | Nordland · Nordland | runtime (s): 0.551 -> 0.036 | Improved |
| use VPR-Cloak instead of MultiANDA | Nordland · Nordland | avg Delta R@1: 59.5 -> 70.9 | Improved |
| use VPR-Cloak instead of MultiANDA on Tokyo247 | Tokyo247 · Tokyo247 | PSNR: 30.20 -> 29.11 | Hurt |
| use VPR-Cloak instead of MultiANDA on MSLS challenge (black-box DHE) | MSLS challenge · MSLS challenge (black-box DHE) | Delta R@1: 61.3 -> 61.6 | No clear effect |
Closest papers by meaning
- PerceptAnon: Exploring the Human Perception of Image Anonymization Beyond Pseudonymization for GDPR (ICML 2024 · similarity 0.85)
- PID: Prompt-Independent Data Protection Against Latent Diffusion Models (ICML 2024 · similarity 0.85)
- Validating Privacy-Preserving Face Recognition under a Minimum Assumption (CVPR 2024 · similarity 0.85)
- MultiPriv: Benchmarking Individual-Level Privacy Reasoning in Vision-Language Models (ICML 2026 · similarity 0.84)
- Do Vision-Language Models Respect Contextual Integrity in Location Disclosure? (ICLR 2026 · similarity 0.84)
- ViP: A Differentially Private Foundation Model for Computer Vision (ICML 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.
