The Human Attention Monitoring System using Webcam is an intelligent real-time application developed to continuously detect and analyse a user's attention level using a standard webcam. The system leverages Computer Vision and Deep Learning techniques to analyse facial features, eye movements, blink rates, and gaze directions in order to determine whether a person is Attentive, Distracted, or Drowsy. This project addresses a critical need in domains such as online education, vehicle driver monitoring, workplace productivity tracking, and remote proctoring. The system is developed using Python, OpenCV, MediaPipe, and Dlib libraries. The webcam captures real-time video frames which are processed through face detection and facial landmark algorithms. Key attention indicators such as Eye Aspect Ratio (EAR), head pose estimation, and gaze direction are calculated from the detectedlandmarks. Based on predefined threshold values, the system classifies the user’s attention state and triggers appropriate visual or audio alerts when distraction or drowsiness is detected. The application also provides a real-time monitoring dashboard displaying attention level graphs, session statistics, and summary reports. The system is lightweight, runs on standard hardware with a regular webcam, and requires no wearable devices or specialised sensors, making it a practical and cost-effective solution for widespread deployment in educational and professional environments.

Keywords — Computer Vision, Eye Aspect Ratio, MediaPipe FaceMesh, Deep Learning, Real-Time Monitoring, Attention Detection, Webcam, Head Pose Estimation