Research map / Image generation / Image restoration
Fingerprinting Deep Image Restoration Models
Yuhui Quan, Huan Teng, Ruotao Xu, Jun Huang, Hui Ji
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 critical-image fingerprints instead of classification-adapted IPGuard baselines | image denoising models · verification under pruning/finetuning/quantization | verification success rate: 67% / 78% -> 100% | Improved |
| use critical-image fingerprints instead of ProjCL on SR models | image SR models · verification under attacks | verification success rate: 0% -> 100% | Improved |
| weaken watermark embedding strength in Quan et al. | watermarking (WeakWM) · verification under attacks | verification success rate: 100% -> 28% (denoising), 100% -> 22% (SR) | Hurt |
| minimize the objective in Eq. 1 rather than maximize it (MaxObj adversarial-like image) (ablation) | critical image fingerprint · SR quantization attack | successful verifications: 3/6 -> 6/6 | Improved |
| include the TV term in the critical image objective (ablation) | critical image fingerprint · denoising and SR models | discriminability | Improved |
| use critical images to fingerprint models of the same architecture trained on different datasets | Restormer · six Restormer variants | pairwise distance vs threshold | Improved |
Closest papers by meaning
- SDBF: Steep-Decision-Boundary Fingerprinting for Hard-Label Tampering Detection of DNN Models (CVPR 2025 · similarity 0.87)
- IPRemover: A Generative Model Inversion Attack against Deep Neural Network Fingerprinting and Watermarking (AAAI 2024 · similarity 0.86)
- Intersecting-Boundary-Sensitive Fingerprinting for Tampering Detection of DNN Models (ICML 2024 · similarity 0.86)
- Fingerprinting Denoising Diffusion Probabilistic Models (CVPR 2025 · similarity 0.86)
- United We Stand, Divided We Fall: Fingerprinting Deep Neural Networks via Adversarial Trajectories (NeurIPS 2024 · similarity 0.84)
- OmniMark: Efficient and Scalable Latent Diffusion Model Fingerprinting (AAAI 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.
