👁️ Drowsiness & Gaze Tracking Web App

A real-time drowsiness and gaze tracking system for online exams or corporate meetings. Built with Streamlit, dlib, OpenCV, and MediaPipe, it provides:

• 🔐 User/admin login system
• 🧠 Eye Aspect Ratio (EAR) based drowsiness detection
• 👀 Gaze monitoring (center, left, right, down)
• 📊 Admin dashboard for session log analysis

---

🚀 Features

👤 User Side

• Live camera monitoring
• EAR + gaze tracking
• Real-time alerts
• Metric logging per session

🛠️ Admin Side

• Upload/browse session logs
• EAR graph over time
• Drowsiness event summary

---

📦 Requirements

pip install -r requirements.txt

Ensure you also have:

• shape_predictor_68_face_landmarks.dat in the root directory
• A working webcam

---

🏃‍♂️ Run the App

streamlit run app.py

---

📁 Project Structure

drowsy_webapp/
├── app.py                # Main routing script
├── login.py              # Login page
├── detection.py          # User webcam detection page
├── dashboard.py          # Admin session dashboard
├── utils/
│   └── session_logger.py # Logs EAR to CSV
├── data/                 # Session logs stored here
└── requirements.txt

---

✅ Demo Credentials

• User: user / 1234
• Admin: admin / admin123

---

🌐 Deployment Ideas

• Streamlit Cloud (easy and free)
• Render / Railway (Flask version)
• Docker + EC2 for production

---

📬 Contact / Ideas

Feel free to extend with:

• Audio alerts 🔊
• Face absence detection 🚫
• Detailed attention scorecard 📈