Facial Recognition Based Attendance System

This project implements an AI-powered attendance management system using real-time facial recognition. The system replaces traditional manual roll-calls and RFID systems with a robust, automated pipeline capable of detecting, recognizing, and recording attendance seamlessly. Built with deep learning models (CNN-based face embeddings and classifiers), it ensures high accuracy even under variations in lighting, pose, or facial features.

The workflow integrates live camera streams with face detection (Haar cascades/SSD/YOLO) and recognition (FaceNet/Dlib embeddings). Recognized individuals are matched with a database and their attendance is logged instantly into structured CSV/SQL databases. The system also generates daily/weekly reports and provides a dashboard view for administrators to monitor attendance trends.

Key advantages include reduced proxy attendance, faster processing, and integration with institutional management systems. Beyond academia, this approach extends to offices, secure workplaces, and industrial compliance scenarios where accurate workforce presence tracking is critical.