Biometric Systems
Redefining identity for the decentralised era — privacy-first recognition systems that never expose biometric templates to a central server.
Biometric Systems
Our biometric research develops advanced multimodal recognition systems using facial, fingerprint, iris, and behavioural biometrics. We prioritise security, privacy preservation, and ethical implementation — using zero-knowledge proofs and federated architectures to ensure biometric data never leaves the device.
Key Research Topics
Multimodal Recognition
Fusing fingerprint, facial, and iris modalities for robust recognition under degraded conditions and partial occlusion.
Iris & Retinal Analysis
Deep learning pipelines for high-resolution iris and retinal biometrics, including analysis under low-light and off-angle conditions.
Privacy-First Architectures
Zero-knowledge proof protocols and homomorphic encryption schemes that verify identity without ever exposing raw templates.
Spoofing & Liveness Detection
Adversarial presentation-attack detection models trained on synthetic and real spoofing datasets to harden recognition pipelines.
Current Projects
ZKBio: Zero-Knowledge Biometric Authentication
A proof-of-concept authentication system using zero-knowledge succinct proofs to verify fingerprint and iris matches without ever exposing raw biometric templates to a server.…
Active · 2024–2026Related Publications
Privacy-Preserving Iris Recognition via Zero-Knowledge Neural Hashing
Elena Rodriguez, Marcus Thorne
IEEE International Conference on Biometrics (ICB)
Gait-Based Identity Verification for Low-Resource Rural Environments
Justice Kwame Appati, David Park
Pattern Recognition Letters
Interested in Collaborating?
We welcome partnerships with institutions, NGOs, and researchers working at the frontier of AI and this domain.