How Blood Flow Exposes Deepfakes
From video capture to liveness verdict — how rPPG turns every camera into a deepfake detector
Simple Steps to Get Started
Measuring your vitals takes less than a minute
Integrate the Circadify SDK
Add Circadify to your existing identity verification or authentication flow with a lightweight SDK. Integration requires no specialized hardware — the SDK works with any standard RGB camera on mobile devices, web browsers, or kiosk systems your users already interact with.
Capture Facial Video During Verification
When a user enters your verification flow, the Circadify SDK captures a brief segment of facial video — typically a few seconds. The user experiences zero additional friction. No blinking challenges, no head movements requested. The capture happens passively during the existing selfie or video step.
Extract Blood Flow Signals via rPPG
On-device algorithms analyze the captured frames for micro-color changes caused by blood pulsing through facial capillaries. Circadify isolates these hemodynamic signals from ambient noise, lighting variations, and compression artifacts — producing a blood volume pulse waveform that only living tissue can generate.
Return a Liveness Verdict
Circadify delivers a definitive liveness result: genuine blood flow detected, or not. Deepfakes, printed photos, screen replays, 3D masks, and virtual camera injections all fail the hemodynamic test. The verdict integrates directly into your decisioning pipeline — approve, escalate, or reject.
See rPPG Liveness Detection in Action
Request an enterprise demo to see how Circadify catches deepfakes that bypass traditional anti-spoofing — using the one biometric signal synthetic media cannot replicate.