Passive Liveness

Passive Liveness

Active Liveness For challenge-response active liveness, see the Active Liveness guide.

Determine if a face image shows a real, live person or a presentation attack (printed photo, screen replay, mask).

How It Works

OpenBiometrics uses passive liveness — no user interaction required. A single image is analyzed for artifacts that indicate spoofing:

• Print attack detection (paper photos)
• Screen replay detection (phones, tablets)
• Texture analysis (mask detection)

Liveness is included in every /detect call automatically.

Usage

Liveness results are part of the face detection response:

{
  "faces": [{
    "liveness": {
      "is_live": true,
      "score": 0.94
    }
  }]
}

Field	Type	Description
is_live	bool	true = real person, false = spoof
score	float	Confidence score 0-1

Best Practices

• Require score > 0.8 for high-security applications
• Combine with quality checks (quality.is_acceptable) for best results
• Use images from a live camera feed, not pre-captured photos
• Ensure adequate lighting for accurate detection