Biometric
Liveness Detection · Anti-spoofing

Liveness Detection API for Face Verification

Confirm that a real person is present during remote identity verification. Add passive or active liveness checks to web and mobile onboarding through API and SDKs.

Start freeView documentation
LIVENESS CHECKPassed
Session photo
Liveness score99.8%
Brightness109
Blur21
Age~25
Checks · No risks detected
Frame 1Frame 2Frame 3Frame 4
<1 secverification time
99.1%accuracy
37 / 570+global ranking
180+countries supported

Clients who trust us

NurbankOtbasy BankBinancenaimi.kzBank RBKTennisi.kzSun FinanceShinhan FinanceKcellNPCK — National Payment Corporation of Kazakhstan
Anti-spoofing

Block 99.9% of biometric attacks of any complexity

* Based on internal testing of 10,000+ attacks

Our AI core analyses not an image, but what cannot be forged: skin texture, micro-expressions, scene depth and lighting changes in real time.

01

Generative deepfakes

Detects AI-generated faces, face-swap masks and software video-stream substitution.

LIVENESS CHECKFailed
Generative deepfakes
Liveness score3.2%
Brightness94
Blur15
Age~28
Checks · Deepfake detected
PhotoVideoDeepfake
02

Video-stream injection attacks

Blocks attempts to feed pre-recorded video through emulators or virtual webcams.

LIVENESS CHECKFailed
Video-stream injection attacks
Liveness score0.1%
Brightness94
Blur15
Age~28
Checks · Stream injection
Stream substitutionVirtual camera
03

Silicone & 3D masks

Detects three-dimensional silicone, latex or wax face casts.

LIVENESS CHECKFailed
Silicone & 3D masks
Liveness score8.7%
Brightness94
Blur15
Age~28
Checks · Mask detected
Mask3D presentation attack
04

Screen & paper replays

Cuts off high-resolution photo/video shown from other screens or printed paper.

LIVENESS CHECKFailed
Screen & paper replays
Liveness score1.4%
Brightness94
Blur15
Age~28
Checks · Screen replay
ScreenPrintout
Scenarios

One module — custom scenarios for any task

When a provider forces a single rigid verification flow, you either lose customers at onboarding or let fraudsters through. With Liveness Core you control the verification logic right in your dashboard (Flow Builder) — no developers needed.

Instant · Fast

Passive mode

A regular selfie is all it takes. The check runs invisibly and instantly.

Regular selfie
Background check
Done in ~1 sec
Perfect for
iGamingSharing servicesE-commerce
Hybrid · Distance

Distance mode

A smart hybrid. The system asks the user to move slightly closer to or further from the screen.

Selfie
Move closer / further
Depth confirmation
Perfect for
Fintech startupsMarketplaces with wallets
Maximum · 3D / PRO

Active mode

A comprehensive check with dynamic actions: turn your head, blink, smile.

Turn your head
Blink
Smile
Perfect for
BankingMicrofinanceContract signing

One API and one module cover any internal compliance rules. Configure retry limits, lighting requirements and confidence thresholds in a couple of clicks.

Set up a scenario for your business
Choosing a mode

Which liveness mode to use

ModeWhat the user doesDepth of the checkWhen to choose itResponse time
PassiveA plain selfie, no movementBaseline: frame analysis and spoofing signalsHigh-volume onboarding where conversion matters~1 second
DistanceMove closer to the camera, then further awayHigher: adds depth-of-scene confirmationFintech and wallets at medium riskDepends on the number of prompts
ActiveTurn the head, blink, smileHighest: dynamic user actionsBanks, microfinance, contract signingDepends on the number of actions

The mode is configured in the Flow Builder: retry count, lighting requirements and the confidence threshold are set per scenario.

LIVENESS CHECKPassed
Session photo
Liveness score99.8%
1Farther
2Closer
Done
Brightness109
Blur21
Age~25
No risks detected1st attempt
UX + Analytics

Complete verification first time — without retries

The system does not just return a pass/fail verdict. It guides the user through verification the first time and collects metadata for security and compliance teams.

Smart hints on the fly
The screen guides the user — “Move your face closer”, “Too dark”, “Remove your glasses”. This minimizes customer drop-off.
Frame and environment quality checks
The model flags closed eyes, masks and extra faces in the frame. Those signals come back with the session result and help you stop an operation when the user is not actually taking part in the check.
Extra analytics
Detection of gender, approximate age and distinctive features — metadata for your security team.
On-Premise deployment

Customer data stays on your servers

Run in our cloud, or deploy the Liveness core into a closed loop on your own servers (On-Premise).

The result: user data never leaves your company perimeter. On-Premise helps you meet local storage and data-control requirements; what applies depends on the jurisdiction and your internal rules.

ISO 27001GDPRNational Bank requirements
Cloud · SaaSOn-Premise
Your company’s perimeter
Liveness core
Customer data
Data never leaves the perimeter
For developers

Classic integration for developers

If your team prefers to build the architecture by hand — use our ready-made SDKs and API.

JS · Webview

Web SDK / Flow Widget

A ready-made JS camera component (widget) for your website or embedded mobile browser (Webview). Fully customizable design to match your brand book.

JavaScriptCustom designWebview
iOS · Android

Mobile SDK

Native libraries for mobile apps. Deep access to the smartphone camera and biometric micro-packet transfer even on weak 3G connections.

SwiftKotlinOffline-ready
Backend

REST API

Direct access to secure liveness-check endpoints for your own server-side logic. Authorization via a private X-API-KEY.

RESTX-API-KEYServer-side
AI · AGENTS

MCP server for AI agents

AI agents connect directly to verification — without writing code. Compatible with Claude, GPT and any MCP client.

MCPClaudeAI agents
liveness — REST API
# Liveness check by session_id
POST /v1/liveness/check/
Host: api.biometric.vision
X-API-KEY: sk_live_••••••••••••••
Content-Type: application/json
{ "session_id": "sess_9f2a…", "mode": "active" }
# → 200 OK
{ "live": true, "score": 0.999, "checks": { "deepfake": "pass", "mask": "pass" } }
FAQ

Questions about liveness checks

In passive mode the user simply looks at the camera and the check runs in the background. In active mode the system asks them to turn their head, blink or smile: the check takes longer but adds dynamic signals of presence. Distance mode sits between the two — it asks the user to move closer or further away and confirms the depth of the scene.

Yes. The Web SDK is a ready-made JavaScript camera component that embeds both in a regular website and in an embedded mobile browser (WebView).

Yes. The Mobile SDK ships as native Swift and Kotlin libraries with direct access to the phone camera.

Yes. Besides the cloud, an On-Premise deployment is available: the core runs on your servers and user data does not leave your perimeter.

The user gets a hint — move closer, too dark, remove glasses — and can try again. Retry count, lighting requirements and the confidence threshold are configurable per scenario, and the session result is visible in the operator cabinet.

Next in the flow: selfie-to-ID face matching · document verification · eKYC for banks

Numbers & facts

433+
Document types —Support for passports and hundreds of other document types worldwide
180+
Supported countries —Global coverage for verification across different jurisdictions
40 sec
Onboarding time —Register and verify a user in under a minute
99%
Technology accuracy —High-precision AI-based document recognition and verification