Facial Recognition Attendance Tracker (FRAT)

A robust, memory-efficient web application designed for automated employee attendance tracking using cutting-edge facial recognition technology. Built with Flask, the system provides a simple web interface for new user registration and a quick, live camera check-in process.

---

✨ Features

Feature	Description
User Onboarding	Secure registration: Name, Employee ID, Face Image upload. 🧑‍💻
Facial Recognition	128D facial encoding (Dlib) generated and stored as a secure binary blob. 💾
Check-In Process	Real-time attendance via web camera; compares live encoding against database. 📸
Data Management	Attendance records (User ID, Date, Time, Status) stored via Flask-SQLAlchemy. 📊
Deployment	Memory-Optimized (Conda/Docker) for constrained cloud environments ($\le$ 512MiB). ☁️

---

💻 Tech Stack

Category	Technology	Purpose
Backend Framework	Python 3.9 (Flask)	Core web application logic and routing.
Facial Recognition	Dlib / face-recognition	Core library for deep learning-based face encoding and comparison.
Image Processing	OpenCV (opencv-python)	Image handling, especially for processing uploaded and captured images.
Database ORM	Flask-SQLAlchemy	Object-Relational Mapper for the SQLite database (database.db).
Production Server	Gunicorn	Production WSGI HTTP server.
Environment/Deployment	Conda / Docker	Used for reliable, memory-efficient installation of binary dependencies.

---

📺 Live Demo:

https://attendance-web-app-14vp.onrender.com/

(Feel free to use a fake face image or any image you like. Please note that all data will be deleted periodically to protect your privacy.)

---

⚙️ Local Setup

Follow these steps to set up the project on your local machine.

Prerequisites

• Python 3.9 or latest version
• Conda (Miniconda or Anaconda) is highly recommended due to the dependencies.

---

Step-by-Step Installation

1. Clone the Repository:

   git clone https://github.com/shivaram-repo/attendance-web-app.git
   cd face-attendance-system

2. Create and Activate Conda Environment:

   conda create -n attendance python=3.9 -y
   conda activate attendance

3. Install Dependencies: You must install the heavy packages (dlib, opencv) via Conda first, and the rest via pip.

   # 3a. Install the heavy binaries from conda-forge
   conda install -c conda-forge dlib=19.24.2 face-recognition opencv numpy -y
   
   # 3b. Install the remaining Python packages (Flask, Gunicorn, etc.)
   pip install -r requirements.txt

4. Run the Application:

   python app.py

The application will start on http://127.0.0.1:5000 (or http://localhost:5000).

---

🐳 Deployment (Dockerized)

The provided Dockerfile uses a robust Conda-based multi-step process to ensure a successful build and a low-memory runtime, which is essential for deployment on platforms with strict resource limits (like Render or Heroku).

1. Build the Docker Image:

   docker build -t face-attendance-app .

2. Run the Container:

   docker run -d -p 8000:8000 --name attendance-server face-attendance-app

   (Note: The app runs on port 8000 inside the container, as configured in the Dockerfile and CMD.)

   ---

Key Deployment Optimizations

• Conda: Used to install resource-heavy packages (dlib, opencv) as pre-compiled binaries, avoiding the C++ compilation causing "Out of memory" errors during the build phase.
• Gunicorn: The command is set to CMD ["gunicorn", "-w", "1", "--threads", "4", "-b", "0.0.0.0:8000", "app:app"] to limit the number of worker processes to one. This aggressively reduces the runtime memory usage, preventing the "Out of memory (used over 512Mi)" failure during startup.

---

🤝 Contribution

Feel free to fork the repository and contribute. For major changes, please open an issue first to discuss what you would like to change.

---

• Note: This project was built by the help of AI and is intended for educational purposes.