TrackEase: Automatic Attendance System Using Face Recognition

---

Project Description

TrackEase is a fully automated attendance management system designed to simplify and enhance the attendance tracking process for educational institutions. Using advanced face recognition models, it enables contactless, real-time attendance marking via live camera feeds. It eliminates manual errors and ensures data accuracy while providing teachers with powerful reporting tools.

---

Features

For Students:

• Attendance Tracking: Attendance is marked automatically when their face is detected in the live feed.
• Attendance History: View attendance history on the platform.

For Teachers:

• Secure Registration: Teachers register via email/phone with OTP verification.
• Real-Time Attendance: Automatically detect and mark student attendance from a live camera feed.
• Attendance Reports: Generate and view reports for specific dates and time periods.

---

Technologies Used

Frontend:

• HTML & CSS + Jinja: For building a fast, responsive, and user-friendly web interface.

Backend:

• Flask: A lightweight Python web framework to handle API requests and backend logic.
• dlib: For face recognition and feature extraction using pre-trained models.
  • dlib_face_recognition_resnet_model_v1.dat: Extracts facial embeddings.
  • shape_predictor_68_face_landmarks.dat: Identifies facial landmarks.

Database:

• MongoDB: For storing user data, facial features, and attendance records.

Other Tools:

• OpenCV: For live camera feed processing.
• DCGAN: For generating synthetic data to augment a small dataset.

---

System Architecture

Frontend (React + Vite) --> Backend (Flask API) --> MongoDB Database
                                      |
                                Face Recognition
                                      |
                             Live Camera Feed (OpenCV)