This project focuses on developing an Automated Classroom Attendance System leveraging Deep Learning and Transformer models to accurately count students in classrooms. By addressing challenges like occlusions, overlapping individuals, and varied lighting conditions, the system ensures high accuracy and robust performance in real-world scenarios.