On the Design of Deep Learning based Facial Recognition System using Smartphone for Class Attendance

Face recognition technology has become a widely used tool in many fields, such as security, authentication, and attendance systems. In this paper, we propose an efficient and scalable real-time face recognition attendance system that utilizes InsightFace for embedding extraction and Qdrant as a high-performance vector database for facial recognition. The system is able to automatically log attendance through image processing and can easily identify registered persons in class or group photographs. It portrays the architecture, design consideration, and implementation of it, with accuracy, performance on a real-time scale, as well as scalability across most practical applications, including educational centers and workplaces.