Presentation Attack Detection (PAD)
Overview
PAD detects attempts to spoof biometric systems using photos, videos, masks, molds, or synthetic signals. It complements liveness detection and is referenced by ISO/IEC 30107-3 testing protocols. Techniques include texture/reflectance analysis, challenge-response, depth sensing, and AI models trained on known attacks.Strong PAD lowers false accepts (FAR) without harming genuine users (FRR) and is essential for remote onboarding and step-up authentication. Regulators increasingly expect PAD for biometric KYC to counter deepfakes and injection attacks. Vendors should publish evaluated performance and undergo third-party testing.
Stay ahead of risk with Signzy
Explore tools that help you onboard, monitor, and verify with confidence
Biometric Verification
Authenticate users with facial, fingerprint, and liveness biometrics powered by AI to prevent identity spoofing and fraud.
One Touch KYC
Launch global KYC flows with built-in document OCR, liveness checks, deepfake detection, and AML, all through a single, customizable dashboard.
Database Verification
Instantly verify user information by connecting to trusted databases across jurisdictions for accurate, compliant, and faster onboarding.
FAQ
How is PAD different from liveness?
Liveness confirms a live subject; PAD focuses on detecting spoofs/presentations crafted to fool sensors. They’re often implemented together.
What methods are most effective?
Multi-modal signals (texture, depth, motion), randomized prompts, and attack-aware AI models improve resilience.
How do we measure quality?
With FAR/FRR and attack presentation classification error rates; independent evaluations build trust.
Where does PAD fail?
Against novel, high-quality spoofs — continuous retraining and attack intel are vital.