TRU.ID
TRU.ID matches a live selfie against the face on an identity document to confirm the person presenting the document is its rightful owner. Combines face match with liveness and deepfake detection to prevent all spoofing attack types.
Why TRU.ID?
- 1:1 face match — high-accuracy AI face recognition comparing selfie to document photo.
- Liveness detection — confirms the selfie is of a live person, not a photo or screen display.
- Deepfake detection — detects AI-generated faces and video injection attacks on the camera feed.
- Privacy-first — no biometric data stored beyond the transaction window.
- RBI-aligned — supports Video KYC (VKYC) and remote identity verification compliance.
Capabilities
1. Face Similarity Score
0–100 confidence score for the selfie-to-document face match. Configurable match threshold per use case.
2. Liveness Detection
Passive and active liveness modes. Detects printed photos, screen replays, and 3D mask attacks.
3. Deepfake Detection
Probability score for AI-generated or video-injected content. Catches GAN faces, face-swap attacks, and diffusion-generated images presented to the camera.
4. Age Estimation
Approximate age from selfie to flag potential identity misuse (e.g., minor using an adult document).
How It Works
- Submit document image and selfie (or video frame) to the TRU.ID endpoint.
- Face match model compares the two images and generates a similarity score.
- Liveness and deepfake detection models run in parallel.
- Returns match verdict, liveness result, deepfake probability, and confidence scores.
Use Cases
1. Remote KYC selfie verification Verify identity for digital lending, NBFC onboarding, and neobank account opening.
2. Video KYC (VKYC) Use TRU.ID as the identity confirmation layer in RBI-compliant VKYC flows.
3. Account recovery re-verification Re-confirm identity for password resets or account unlocking requests.
4. Step-up authentication Trigger a face match challenge when a transaction exceeds risk thresholds.
View API Reference →