Selfie and liveness detection

Selfie and liveness detection uses AI-powered analysis to confirm that the person completing verification is physically present and that their face biometrically matches the photo on their government-issued ID.

This check is designed to prevent fraud by ensuring a real person is behind the transaction and that they are the legitimate owner of the identity document. By analyzing biometric data alongside 900+ unique fraud signals that span across digital footprint, device intelligence, behavioral biometrics and IP reputation, SEON can quickly determine not only whether the user is live and matches their ID, but also their true intent.

The selfie and liveness detection is a passive, low friction user-friendly verification flow that adds a critical layer of security to the identity verification journey.

1. Selfie capture initiation: The user is prompted to take a selfie as part of the verification workflow.

2. Guided selfie capture: The SEON SDK provides a camera-guided interface to help the user frame their face correctly. The liveness check is performed passively in the background without requiring any specific user actions.

3. User intent analysis: The captured selfie is sent to SEON’s servers for analysis. In parallel, SEON’s fraud engine gathers hundreds of data points — including device intelligence, IP data and digital footprint information — to analyze the context of the verification. The system performs a liveness check to prevent spoofing and a vector-based face match against the photo on the ID document.

4. Unified verification results: A final result (Pass, Fail or Review) is returned with a detailed breakdown of all checks performed. This unified outcome, enriched with fraud scores and signals about user intent, is delivered via webhook and is available in SEON.

SEON performs a comprehensive set of checks to ensure the user is physically present and is the legitimate owner of the identity document.

• Passive liveness: Verifies the user is a live person by analyzing subtle cues in the selfie image. This defends against spoofing attempts using photos, videos or masks, without requiring the user to perform any actions.
• Anti-spoofing detection: Actively looks for signs of presentation attacks, such as screen captures, printed photos and other common spoofing techniques.
• Face match: A vector-based comparison that biometrically matches the user's selfie against the face extracted from their government-issued ID document or a previously stored reference image.
• Confidence scoring: Provides a detailed score indicating the confidence level of both the liveness check and the face match, allowing for granular risk-based rules.

During the selfie verification process, the SEON SDK provides real-time instructions to ensure optimal face positioning and liveness detection accuracy:

These prompts guide users through the liveness check process, ensuring high-quality selfie capture for accurate face matching and anti-spoofing detection.

As part of the verification process, SEON automatically extracts the selfie image and creates a biometric template (vector) for future comparisons. This enables faster and more secure re-authentication.

Extracted data includes:

• High-quality selfie image
• Biometric face vector
• Liveness confidence score
• Face match score

• Secure customer onboarding: Add a mandatory liveness and face match check during onboarding to ensure the person creating the account is the same person on the ID document.
• High-value transaction verification: For high-risk actions like large withdrawals, require a quick selfie check to re-authenticate the user and prevent account takeover fraud.
• Password resets and account recovery: Ensure the legitimate account owner is the one requesting to reset their credentials by requiring a selfie verification.
• Step-up authentication: Use a selfie check as a step-up verification method when a user's activity is flagged as suspicious by the fraud engine.

Selfie and liveness verification is initiated and managed through SEON’s mobile and web SDKs. The SDKs provide the complete user interface for guiding users through the selfie capture process, ensuring high-quality images and a frictionless experience.

To start a verification, your application will initialize the SEON SDK and configure it for a selfie verification session. The SDK handles the camera interface, user guidance and secure submission of images to SEON’s servers.

For detailed instructions on implementing the SDKs, please refer to the iOS, Android and Web SDK developer guides.

• Standalone selfie with face match: For re-authentication, a selfie check can be run against a previously stored reference image from an earlier verification session without requiring a new document scan.
• Device requirements: Requires a device with a functioning front-facing camera.