Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{179890,
        author = {G.Manisha and Busala Yashwanth Kumar and E. Karthik Sai and C. Nikhil and Mr. K. Suresh Babu},
        title = {AI-Powered Student Monitoring for Smart Classrooms using ML},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {12},
        pages = {8622-8630},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=179890},
        abstract = {Student behavior plays a vital role in 
improving and maintaining student involvement and 
discipline for better academic outcomes. The current 
study proposes an integrated framework that detects 
activities like sleeping, eating, and using mobiles during 
class hours and notifies the faculty in real time. It also 
tracks student attendance. The integration of Artificial 
Intelligence has become essential in classrooms. This 
system integrates computer vision algorithms, which 
includes YOLOv8 for object detection, MediaPipe for 
facial and hand movement recognition and Twilio 
WhatsApp API, which sends notifications directly to 
the faculty whenever a student is detected doing any 
inappropriate activity like above mentioned activities. 
In this study, we collected classroom footage and 
applied techniques like frame filtering, face alignment 
and activity annotation. We compared deep learning 
models like YOLOv7, YOLOv8 and EfficientDet for the 
highest performance; among these, YOLOv8 has the 
highest detection accuracy of 94.87%, recall of 95.76% 
and an F1-score of 94.01%. Face recognition models 
like FaceNet and Dlib are used for attendance tracking, 
with FaceNet achieving an accuracy of 96.45%. 
Compared to old manual monitoring techniques, our 
system delivers better accuracy and enables faculty 
response faster. This system covers the gap in 
traditional classroom monitoring and provides an 
innovative and automatic solution to improve student 
discipline and engagement. Supporting swift action 
while lightening the load of faculty, this solution 
combines deep learning and real-time notifications. 
This research makes a noteworthy contribution to the 
development of intelligent educational environments 
and encourages the use of ethical AI to develop more 
accurate smart analysis approach models on data up to 
October 2023.},
        keywords = {Computer vision, Deep learning, Real-time  notifications, FaceNet, MediaPipe.},
        month = {June},
        }