ABSTRACT: Attendance recording is an important administrative activity in educational institutions, but manual roll calls and registers consume lecture time and may be affected by human error, proxy attendance, and fragmented record keeping. This paper presents the design and ongoing development of an Intelligent Face Recognition Based Attendance Management System intended to connect face-based student identification with academic structure, timetable scheduling, lecture sessions, and attendance reporting. The proposed application uses Angular for the web interface and Django with Django REST Framework for core application logic and APIs. MySQL is planned for persistent relational data, while Redis is intended for caching and temporary state. Face-recognition processing is designed as a separate Python service using FastAPI, with InsightFace/ArcFace-based embeddings and FAISS-based similarity search proposed for identification. The system is planned to support role-based access, student and teacher records, subject offerings, timetable management, face enrollment, lecture-wise attendance, teacher review, and reports. Since the project is currently under development, this paper documents the problem, related concepts, proposed architecture, methodology, and evaluation plan. It does not claim completed experiments or measured recognition performance. The next stage will assess recognition quality, false acceptance and rejection, processing latency, and system behaviour under realistic classroom conditions.
KEYWORDS : Face Recognition, Attendance Management, Artificial Intelligence, Face Embeddings, ArcFace, FAISS, Django REST Framework, Academic Information Systems.
INTELLIGENT FACE RECOGNITION BASED ATTENDANCE MANAGEMENT SYSTEM
KAMAL K. MISTRY