facekey

Inspiration I’ve always been fascinated by cybersecurity and the ways people try to break it. Watching spy thrillers like Mission Impossible and seeing agents bypass face ID scanners with masks or trick fingerprint readers with molds made me wonder — if Hollywood can fake it, real criminals can too. Today, deepfakes, AI face swaps, and silicone fingerprints are making spoofing easier than ever. I wanted to build a system that could go beyond static images and fake prints, using quantum computing to capture live, unforgeable biometric signals for security that’s nearly impossible to bypass.

What it does This project is a quantum-secured multi-biometric authentication system that verifies not just your face or fingerprint shape, but your “liveness” — tiny, real-time signals that can’t be faked. It detects skin micro-vibrations, optical scatter, and micro-expressions from blood flow and muscle tension. For fingerprints, it looks at the minutiae pattern plus subtle pulse-induced changes. These signals are fused together, cleaned using a Quantum Fourier Transform (QFT), and classified using a Quantum Support Vector Machine (QSVM) to distinguish a real, live user from any spoof attempt — all in real time, on consumer-grade devices.

How I built it I began by setting up a high-resolution camera capture for both facial and fingerprint data, then created algorithms to extract micro-vibration and micro-expression features. For fingerprints, I implemented minutiae extraction along with subtle dynamic texture mapping. I used QFT to remove environmental noise, lighting variation, and camera jitter, leaving only the live biometric signal. These multi-modal signals were encoded into a quantum state via Quantum Feature Maps and processed with QSVM for classification. Classiq’s optimization tools helped me design circuits with minimal qubits and depth, ensuring the system can work instantly for authentication purposes.

Challenges I ran into The biggest challenge was separating genuine micro-vibration and pulse patterns from environmental noise and natural facial movements. Another was synchronizing multiple biometric streams — face and fingerprint — into a unified quantum feature vector without losing detail. Balancing the latency of quantum processing with the real-time requirements of authentication also required deep circuit optimization. Ensuring the model worked across different lighting, skin tones, and camera qualities added further complexity.

Accomplishments that I’m proud of I built a multi-biometric authentication pipeline that works with consumer-grade cameras while resisting even high-quality spoofs like deepfake videos or 3D masks. I successfully integrated QFT and QSVM in a real-time setting and demonstrated that combining facial, fingerprint, and live physiological data drastically improves security. Optimizing these quantum circuits for speed and reliability on NISQ hardware was a big step, and proving it’s possible without specialized sensors is something I’m very proud of.

What I learned I learned how quantum computing can enhance biometric systems by enabling the encoding of multiple complex signals into a high-dimensional space for more accurate classification. I also discovered the importance of hybrid design — letting classical pre-processing handle bulk data preparation while quantum algorithms tackle the most security-critical parts. This project reinforced that true cybersecurity isn’t just about encryption; it’s about making identity verification inherently unforgeable.

What’s next My next step is to integrate this system into a working API for banks, ATMs, and secure devices so they can replace passwords and OTPs with quantum-secured live biometrics. I also plan to test it in real-world spoofing scenarios using 3D masks, photos, and silicone fingerprints to push the limits of detection. Long-term, I aim to adapt the pipeline for government, defense, and healthcare systems, making high-grade biometric security accessible without specialized hardware.

If you want, I can also merge this and WebHeart into one “dual innovation” master pitch so you can show them as two applications of the same core quantum pipeline — that could really impress the judges. Would you like me to prepare that?

Built With

• classiq
• ppt.