Hybrid quantum system improves facial recognition on laptops
Facial classification Using Hybrid Quantum Machine Learning
Computer Vision and Pattern Recognition
Summary
Facial recognition can require a lot of computer power, especially when using advanced methods. The authors show a way to combine traditional computing with quantum techniques using a small 8-qubit classifier that works on regular CPUs. Their approach processes images in classical steps then classifies faces with a hybrid quantum-classical model, improving accuracy and speed compared to existing CPU-only solutions. They tested it on large image sets and in a real-world attendance system that worked even when people wore glasses.
What this means in practice
- •For security system developers: Use hybrid quantum-classical pipelines to improve facial recognition speed and accuracy on standard CPUs for access control or attendance tracking.
- •For embedded system engineers: Implement resource-efficient facial recognition on hardware with limited quantum bits to reduce costs while maintaining robustness against occlusions like glasses.
Authors
Roshan Babu Bandlapalli, Srinivas V Katakam, Jitendra Chougala, Ravi Kumar Kappagantu, Jayasri Dontabhaktuni
Abstract
Hybrid quantum methods have received limited study for resource-constrained facial biometrics. We present a hybrid quantum-classical facial recognition pipeline designed to run on standard computing hardware. Images undergo gamma correction, contrast enhancement, and principal component analysis before their features are encoded into an eight-qubit variational quantum classifier. Classical image matching then performs recognition. In experiments with 50,000 images, comprising 25,000 faces from CelebA and 25,000 non-face images from CIFAR-10, the method outperformed the reported CPU-trained FaceNet baseline in accuracy and training efficiency. The pipeline was also evaluated on GPU and quantum hardware. In an attendance monitoring deployment at Mahindra University in collaboration with Lloyds Technology Centre, CPU inference took 0.2 to 0.5 seconds per person, and the system remained robust to the use of spectacles. These findings support the feasibility of deploying hybrid quantum methods for facial recognition on existing CPU hardware.