Researchers have developed a new method for detecting fake fingerprints by analyzing regions with varying levels of artifact extraction difficulty (AED). Their proposed Cascade Forgery Mining Network (CFM-Net) uses local Gabor feature certainty to partition fingerprint images and adaptively extracts features from these regions. An Orientation Guided Adversarial Training (OGAT) module is also introduced to preserve artifact evidence while filtering out identity information. Experiments on LivDet datasets show CFM-Net outperforms existing methods, particularly on fingerprints with high AED. AI
IMPACT This research could enhance the security of biometric systems by improving the accuracy of detecting sophisticated fingerprint spoofing attempts.
RANK_REASON The cluster contains an academic paper detailing a new method for a specific computer vision task. [lever_c_demoted from research: ic=1 ai=1.0]
- arXiv
- Cascade Forgery Mining Network
- CFM-Net
- Fingerprint Presentation Attack Detection
- Gábor
- LivDet
- Ogata
- Orientation Guided Adversarial Training
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
