Research map / Image generation / Autoregressive and latent generation
One for All: Synthesis-Free Fingerprint Learning for Attribution of In-the-Wild Synthetic Images
Jianwei Fei, Yunshu Dai, Peipeng Yu, Zhihua Xia, Dasara Shullani, Daniele Baracchi, Alessandro Piva
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 the synthesis-free fingerprint extractor instead of the prior Siamese attribution method | fingerprint extractor · DMDetection | average AUC: 37.93 -> 94.05 | Improved |
| use the synthesis-free fingerprint extractor instead of the prior Siamese attribution method | fingerprint extractor · AIGCBenchmark | average AUC: 48.89 -> 83.05 | Improved |
| use the fingerprint extractor instead of CLIP for cross-dataset StyleGAN verification | fingerprint extractor · StyleGAN2/3 SG2-ffhq vs SG3-ffhq | AUC: 30.48 -> 91.07 | Improved |
| use the fingerprint extractor instead of UA(LP+) on GenImage | fingerprint extractor · GenImage | OSCR: 80.78 -> 85.08 | Improved |
| use the fingerprint extractor instead of UA(LP+) on GenImage | fingerprint extractor · GenImage | accuracy: 97.82 -> 96.87 | No clear effect |
| use the fingerprint extractor instead of POSE on OSMA | fingerprint extractor · OSMA average | OSCR: 78.75 -> 88.48 | Improved |
| use the fingerprint extractor instead of POSE on OSMA unseen architecture | fingerprint extractor · OSMA unseen architecture | OSCR: 81.62 -> 82.20 | No clear effect |
| use the fingerprint extractor instead of POSE on OSMA closed set | fingerprint extractor · OSMA | accuracy: 94.81 -> 98.67 | Improved |
Closest papers by meaning
- FingerprintNet: Synthesized Fingerprints for Generated Image Detection (ECCV 2022 · similarity 0.93)
- PrintsGAN: Synthetic Fingerprint Generator (TPAMI 2022 · similarity 0.91)
- ManiFPT: Defining and Analyzing Fingerprints of Generative Models (CVPR 2024 · similarity 0.90)
- WOUAF: Weight Modulation for User Attribution and Fingerprinting in Text-to-Image Diffusion Models (CVPR 2024 · similarity 0.88)
- OmniMark: Efficient and Scalable Latent Diffusion Model Fingerprinting (AAAI 2025 · similarity 0.87)
- Forensic Self-Descriptions Are All You Need for Zero-Shot Detection, Open-Set Source Attribution, and Clustering of AI-generated Images (CVPR 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.
