FaceAttend: A Web-Based Face Recognition Attendance System

Abstract

Traditional student attendance procedures continue to present several challenges, including inefficient processing times, susceptibility to recording errors, and opportunities for fraudulent practices such as proxy attendance. To address these limitations, this study proposes and develops a web-based attendance management system that utilizes facial recognition technology for automatic identity verification. The system was implemented using React for the user interface, FastAPI for backend services, Supabase for data storage and management, and the InsightFace Buffalo_S model based on the ArcFace architecture for facial recognition.

The development process involved facial image acquisition, feature embedding generation, system integration, and performance evaluation of the recognition model. Experimental results demonstrated that the proposed system was able to identify registered users with a high level of reliability, achieving 100% accuracy, 100% precision, and 100% recall on the testing dataset. Furthermore, both the False Acceptance Rate (FAR) and False Rejection Rate (FRR) were recorded at 0%, indicating the absence of recognition and verification errors during testing.

These findings suggest that the proposed facial recognition attendance system can serve as a practical, secure, and efficient solution for automating student attendance tracking while minimizing the risk of attendance fraud in academic environments.