Abstract:
This study presents an automated attendance system using face recognition and deep learning,
improving efficiency by detecting students in real time and reducing manual tracking. Optimizations
like multi-threading and frame skipping enhance speed and accuracy, supporting up to 10 faces
simultaneously. Attendance records are stored in CSV/Excel, while a PyQt-based GUI ensures userfriendly interaction for student registration, automated tracking, and administrator control. The
system reduces errors, increases reliability, and can be adapted for educational institutions,
workplaces, and security systems.
