Liveness Detection

Liveness detection answers a narrow but critical question: is the biometric sample coming from a real, living person at this moment? Without it, a facial recognition or fingerprint system can be fooled by a printed photo, a replayed video, a 3D mask, or a lifted print.

Techniques fall into two groups. Passive methods analyze a single capture for signs of life, such as skin texture, reflection, or depth, without asking the user to act. Active methods prompt a challenge, such as turning the head or blinking, and confirm the response. The international reference for evaluating these defenses is ISO/IEC 30107, which standardizes presentation attack detection testing.

Liveness is what makes remote, unsupervised identity verification trustworthy. It is the difference between confirming a document matches a face and confirming that face belongs to someone actually present.

For CIAM, liveness detection underpins remote onboarding and KYC: it lets a business accept a self-captured selfie or video as evidence of a real applicant while resisting the spoofing attacks that target unsupervised enrollment.

Sources

• ISO/IEC 30107-3:2023, Biometric presentation attack detection: https://www.iso.org/standard/79520.html