Research map / Image generation / Autoregressive and latent generation
PrintsGAN: Synthetic Fingerprint Generator
Joshua J. Engelsma, Steven A. Grosz, Anil K. Jain
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 |
|---|---|---|---|
| pretrain on PrintsGAN then finetune on N2N instead of N2N alone | DeepPrint · NIST SD4 | TAR @ 0.01% FAR: 73.37 -> 87.03 | Improved |
| pretrain on PrintsGAN then finetune on N2N instead of N2N alone | DeepPrint · FVC 2002 DB1 A | TAR @ 0.01% FAR: 79.68 -> 89.74 | Improved |
| pretrain on PrintsGAN then finetune on N2N instead of N2N alone | DeepPrint · FVC 2004 DB1 A | TAR @ 0.01% FAR: 65.99 -> 90.22 | Improved |
| pretrain on Sfinge then finetune on N2N instead of N2N alone | DeepPrint · NIST SD4 | TAR @ 0.01% FAR: 73.37 -> 54.70 | Hurt |
| pretrain on Sfinge then finetune on N2N instead of N2N alone | DeepPrint · FVC 2002 DB1 A | TAR @ 0.01% FAR: 79.68 -> 56.68 | Hurt |
| use PrintsGAN instead of Sfinge synthetic training data only | DeepPrint · NIST SD4 | TAR @ 0.01% FAR: 10.78 -> 52.65 | Improved |
| use PrintsGAN instead of Sfinge synthetic training data only | DeepPrint · FVC 2004 DB1 A | TAR @ 0.01% FAR: 20.80 -> 22.35 | Improved |
| add PrintsGAN pretraining to N2N training | DeepPrint · NIST SD4 closed-set identification | rank-1 identification rate: 85.90 -> 92.05 | Improved |
| add PrintsGAN pretraining to N2N training | DeepPrint · NIST SD4 + 100k real gallery | rank-1 identification rate: 73.20 -> 86.80 | Improved |
| use PrintsGAN rolled prints instead of Mistry et al. rolled prints | synthetic generator · Expert realism ranking | average realism rating: 3.70 -> 2.45 | Improved |
Later papers that compared with this method
1 later accepted paper reports PrintsGAN as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Universal Fingerprint Generation: Controllable Diffusion Model With Multimodal Conditions (TPAMI 2024 · better on 6 of 6 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- One for All: Synthesis-Free Fingerprint Learning for Attribution of In-the-Wild Synthetic Images (AAAI 2026 · similarity 0.91)
- Scalable Fingerprinting of Large Language Models (NeurIPS 2025 · similarity 0.87)
- OmniMark: Efficient and Scalable Latent Diffusion Model Fingerprinting (AAAI 2025 · similarity 0.86)
- Blind-Touch: Homomorphic Encryption-Based Distributed Neural Network Inference for Privacy-Preserving Fingerprint Authentication (AAAI 2024 · similarity 0.84)
- ManiFPT: Defining and Analyzing Fingerprints of Generative Models (CVPR 2024 · similarity 0.84)
- Fingerprinting Denoising Diffusion Probabilistic Models (CVPR 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.
