Face Authentication Attendance System

A robust, Python-based Face Authentication System for tracking attendance. It features real-time face recognition, liveness detection to prevent photo spoofing, and a modern GUI.

🚀 Features

• Two Interface Modes:
  • Desktop App: Modern Dark Mode GUI (via CustomTkinter) for high-performance, real-time logging.
  • Web Dashboard: Lightweight Web App (via Streamlit) for easy access and mobile testing.
• Face Recognition: Uses VGG-Face (Deep Learning) for high-accuracy identification.
• Liveness Detection: Blink detection to ensure the user is present (prevents holding up a photo).
• Attendance Logging: Automatically logs "Punch In" and "Punch Out" events with timestamps to CSV.
• One-Shot Registration: Instantly register new users without re-training the model.

🛠️ Technology Stack

• Language: Python 3.12+
• Computer Vision: OpenCV (Haar Cascades)
• AI Model: VGG-Face (via deepface)
• GUI: CustomTkinter
• Web Framework: Streamlit

📦 Installation

1. Clone the Repository and enter the directory.
2. Create a Virtual Environment (Recommended):

   python -m venv venv
   # Windows:
   .\venv\Scripts\activate
   # Mac/Linux:
   source venv/bin/activate

3. Install Dependencies:

   pip install -r requirements.txt

🖥️ Usage

Option 1: Desktop Application (Recommended)

This is the main application with real-time video feedback and blink counting.

python main.py

Controls:

• Register Face: Enter a name in the text box and click "Register Face".
• Punch In: Click the Green button. (Requires looking at the camera and blinking).
• Punch Out: Click the Red button.
• Logs: View recent logs instantly in the text panel.

Option 2: Web Application

A browser-based version suitable for quick checks or remote access.

python -m streamlit run app.py

📄 Documentation

For a detailed technical explanation of the models, algorithms, and failure cases, please refer to the Technical Report (PDF) included in this repository.

📂 Project Structure

• main.py: Entry point for the Desktop GUI App.
• app.py: Entry point for the Streamlit Web App.
• face_auth.py: Core logic for Face Recognition (DeepFace).
• liveness.py: Logic for Blink Detection.
• data/: Stores registered face embeddings.
• logs/: Stores daily CSV attendance logs.