Biometric Search (1:N search) using Fully Homomorphic Encryption (FHE)
This demo shows Suraksh.AI's biometric search solution under FHE.
Scenario 1: Searching an enrolled subject. For this scenario, the reference and probe should be from the same subject. Expected outcome: ✔️ Found
Scenario 2: Searching an enrolled subject with high recognition threshold. For this scenario, the reference and probe should be from the same subject and the recognition threshold set to a high value. Expected outcome: ❌ Not Found
Scenario 3: Searching a non-enrolled subject. For this scenario, choose a probe not enrolled. Expected outcome: ❌ Not Found
Scenario 4: Searching a non-enrolled subject with low recognition threshold. For this scenario, choose a probe not enrolled and lower the high recognition threshold. Expected outcome: ✔️ Found
Setup Phase: 🔐 Generate the FHE public and secret keys.
Choose a security level
Phase 1: Enrollment
Step 1: Upload or select a reference facial image for enrollment.
Step 2: Generate reference embedding.
Choose a face recognition model
Facial embeddings are INVERTIBLE and lead to the RECONSTRUCTION of their raw facial images.
Example:
Facial embeddings protection is a must! At Suraksh.AI, we protect facial embeddings using FHE.
Step 3: 🔒 Encrypt reference embedding using FHE.
Set subject offset.
Step 4: 🔒 Add encrypt reference to the encrypted reference DB.
Phase 2: Search
Step 1: Upload or select a probe facial image for search.
