Research map / Image generation / Autoregressive and latent generation
RaspiReader: Open Source Fingerprint Reader
Joshua J. Engelsma, Kai Cao, 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 |
|---|---|---|---|
| use RaspiReader Direct images instead of COTS_A grayscale images | CLBP/LBP + SVM · Spoof detection, known-material (5-fold) | TDR @ FDR=1.0%: 75.9% -> 98.10% | Improved |
| use RaspiReader raw FTIR images instead of COTS_A grayscale images | CLBP/LBP + SVM · Spoof detection, known-material (5-fold) | TDR @ FDR=1.0%: 75.9% -> 91.5% | Improved |
| add feature-level fusion of raw FTIR and Direct CLBP features (ablation) | CLBP + SVM · Spoof detection, known-material (5-fold) | TDR @ FDR=1.0%: 98.10% -> 97.7% | No clear effect |
| use RaspiReader processed FTIR images instead of COTS_A images | MobileNet · Spoof detection, known-material (5-fold) | TDR @ FDR=1.0%: 91.9% -> 94.5% | Improved |
| use RaspiReader raw FTIR images instead of COTS_A images | MobileNet · Spoof detection, known-material (5-fold) | TDR @ FDR=1.0%: 91.9% -> 95.1% | Improved |
| add score fusion of raw FTIR and Direct models (Fusion 2) (ablation) | MobileNet · Spoof detection, known-material (5-fold) | TDR @ FDR=1.0%: 95.3% -> 98.4% | Improved |
| add processed FTIR model to the score fusion (Fusion 3) (ablation) | MobileNet · Spoof detection, known-material (5-fold) | TDR @ FDR=1.0%: 98.4% -> 98.9% | Improved |
| replace COTS_A images with RaspiReader fusion for unseen Ecoflex | CLBP/LBP + SVM · Spoof detection, cross-material (unseen Ecoflex) | TDR @ FDR=1.0%: 66.0% -> 77.0% | Improved |
| replace COTS_A images with RaspiReader fusion for unseen Wood Glue | CLBP/LBP + SVM · Spoof detection, cross-material (unseen Wood Glue) | TDR @ FDR=1.0%: 100.0% -> 81.0% | Hurt |
| replace COTS_A images with RaspiReader Fusion 3 for unseen Crayola Model Magic | MobileNet · Spoof detection, cross-material MobileNet (unseen Crayola Model Magic) | TDR @ FDR=1.0%: 50.0% -> 100.0% | Improved |
| replace COTS_A images with RaspiReader Fusion 2 for unseen Ecoflex | MobileNet · Spoof detection, cross-material MobileNet (unseen Ecoflex) | TDR @ FDR=1.0%: 100.0% -> 8.0% | Hurt |
| add processed FTIR model to Fusion 2 for unseen Ecoflex (ablation) | MobileNet · Spoof detection, cross-material MobileNet (unseen Ecoflex) | TDR @ FDR=1.0%: 8.0% -> 56.0% | Improved |
Closest papers by meaning
- PrintsGAN: Synthetic Fingerprint Generator (TPAMI 2022 · similarity 0.83)
- FingerprintNet: Synthesized Fingerprints for Generated Image Detection (ECCV 2022 · similarity 0.81)
- OmniMark: Efficient and Scalable Latent Diffusion Model Fingerprinting (AAAI 2025 · similarity 0.79)
- Scalable Fingerprinting of Large Language Models (NeurIPS 2025 · similarity 0.79)
- ForensicHub: A Unified Benchmark & Codebase for All-Domain Fake Image Detection and Localization (NeurIPS 2025 · similarity 0.77)
- SIF: Semantically In-Distribution Fingerprints for Large Vision-Language Models (CVPR 2026 · similarity 0.77)
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.
