A Deep Learning-Based Face Recognition Approach for Automated Student Attendance Management
A Deep Learning-Based Face Recognition Approach for Automated Student Attendance Management
Mr.M BHANU PRAKASH 1
Assistant Professor, Department ofAI&DSAnnamacharya Institute ofTechnology and Sciences, Tirupati – 517520, A.P.
mmbaluprakash@gmail.com
D VARSHITHA 4
UG Scholar, Department ofAI&DSAnnamacharya Institute ofTechnology and Sciences, Tirupati – 517520, A.P.
damarakkupamv@gmail.com
A POOJASREE 2
UG Scholar, Department ofAI&DSAnnamacharya Institute ofTechnology and Sciences, Tirupati – 517520, A.P.
apsreddy10@gmail.com
V YAMINI 5
UG Scholar, Department ofAI&DSAnnamacharya Institute ofTechnology and Sciences,Tirupati – 517520, A.P.
vyshnaviyaminimadhu@gmail.com
A MOUNUSHA 3
UG Scholar, Department ofAI&DSAnnamacharya Institute ofTechnology and Sciences, Tirupati – 517520, A.P.
amounushamounu@gmail.com
Abstract—In most educational institutes, the attendance monitoring is either manual or prone to human errors, as there is more scope for proxy attendance. This work proposes a deep learning-based face recognition methodology that can automate the process of student attendance monitoring. The proposed system inputs real time video frames through a camera feed. The MTCNN identifies the face from detected video frames. The localised facial regions are then pre- processed and embedded into a compact and discriminative facial embedding using the deep learning model. Similarity-based matching of these embeddings against stored student encoding enables accurate identification. Further, after the confirmation of a match, automatic logging of attendance will be recorded in the database with details about date and time. In addition, this work develops a web-based platform based on the Django framework to provide all kinds of facilities solution provides a reliable contactless scalable attendancemonitoring approach which is suitable for academic environments. The study concludes: Automated Attendance Monitoring,Face Recognition, Deep Learning, Multi-task Cascaded Convolutional Networks, Facial Embeddings, Computer Vision, Django Framework.