Liveness Detection proves the face in front of the camera belongs to a real, present human — not a photo, video replay, silicon mask, or deepfake. KYC.bd ships both passive (the user does nothing, we analyze a 1.5-second clip) and active (the user blinks or turns) modes from a single SDK.
Anti-spoofing coverage
- Print attack: Photo printed on paper or glossy media.
- Replay attack: Video played on a phone, tablet, or laptop.
- 3D mask: Silicone, latex, and resin masks.
- Deepfake: Real-time face-swap from open-source and commercial tools.
Our presentation-attack detection (PAD) is tested against the ISO/IEC 30107-3 protocol and benchmarked quarterly against the iBeta Level 1 + 2 test sets.
SDK or browser
Web SDK (WebAssembly, ~600KB gzipped), Android SDK (~2MB AAR), iOS SDK (~3MB framework). All three return the same structured score so your backend logic is uniform.
Pairs with NID verification. See the insurance solution for a typical claims-stage deployment.