Research map / Federated learning and privacy / Differential privacy
LDP-Slicing: Local Differential Privacy for Images via Randomized Bit-Plane Slicing
Yuanming Cao, Chengqi Li, Wenbo He
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 uniform privacy budget with utility-aware optimized allocation (ablation) | LDP-Slicing · AgeDB-30 (eps_total=20) | accuracy (%): 89.82 -> 96.68 | Improved |
| replace uniform budget with optimized allocation (ablation) | LDP-Slicing · CPLFW (eps_total=20) | accuracy (%): 86.73 -> 91.08 | Improved |
| use DWT LL pruning instead of DCT-based DC pruning (ablation) | LDP-Slicing · AgeDB-30 (eps_total=20) | accuracy (%): 95.58 -> 96.68 | Improved |
| add LL pruning (perceptual obfuscation) (ablation) | LDP-Slicing · AgeDB-30 (eps_total=20) | accuracy (%): 96.90 -> 96.68 | Hurt |
| add LL pruning (ablation) | LDP-Slicing · LFW / CALFW (eps_total=20) | accuracy (%): 99.77 -> 99.75 / 96.05 -> 96.02 | No clear effect |
| increase total privacy budget from 5.2 to 20 (ablation) | LDP-Slicing · AgeDB-30 | accuracy (%): 68.89 -> 96.68 | Improved |
| increase total privacy budget from 20 to 58 (ablation) | LDP-Slicing · AgeDB-30 | accuracy (%): 96.68 -> 97.45 | Improved |
| use LDP-Slicing instead of DCTDP | face recognition privacy · LFW identity distinguishing attack | attacker advantage (%): 12.58 -> 4.5 | Improved |
| use LDP-Slicing instead of PEEP | face recognition privacy · AgeDB-30 | accuracy (%): 87.47 -> 96.68 | Improved |
Closest papers by meaning
- Validating Privacy-Preserving Face Recognition under a Minimum Assumption (CVPR 2024 · similarity 0.86)
- PID: Prompt-Independent Data Protection Against Latent Diffusion Models (ICML 2024 · similarity 0.86)
- Certification for Differentially Private Prediction in Gradient-Based Training (ICML 2025 · similarity 0.86)
- Differentially Private Representation Learning via Image Captioning (ICML 2024 · similarity 0.86)
- Towards hyperparameter-free optimization with differential privacy (ICLR 2025 · similarity 0.86)
- Gaussian Differentially Private Human Faces Under a Face Radial Curve Representation (ICLR 2025 · similarity 0.86)
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.
