Paper Title
Deep Learning-Powered Web Attendance Tracker

Abstract
In an era where technology plays an ever-increasing role in education, the need for efficient attendance monitoring systems has become paramount. In response to this, we present our ambitious project: a web-based attendance monitoring system that harnesses the power of face recognition algorithms to enhance attendance tracking in educational institutions. The main objective of our project is to compare and identify the most accurate face recognition algorithm among many distinct contenders. Leveraging a diverse and comprehensive database containing numerous student photos captured under varying conditions such as different lighting, angles, and even facial hair, our system aims to learn and adapt to real-world scenarios. For this, we will be comparing 2 algorithms which are- MTCNN with Face Net and Deep Face. To achieve our goals, we have meticulously planned a two-fold approach. Initially, we will test our system using a dummy dataset, allowing us to fine-tune and optimize its performance. Subsequently, we will manually create an authentic database with information about actual students, ensuring a robust and reliable system that caters to real-world needs. Additionally, our system will play a pivotal role in fostering transparent and effective communication between educational institutions and parents. By notifying parents about their ward’s attendance records, we aim to bridge the information gap and encourage parental involvement in their child’s academic journey. Keywords: Attendance tracking, Deep Face, Face Net, MTCNN.