👤 Face Recognition System

AI-Powered Face Recognition with Dual-Panel Login

Live Demo · API Docs

---

✨ Overview

A production-ready face recognition system with separate portals for Admins and Users. Admins can register faces and manage users; Users can perform real-time face recognition. Built with FastAPI backend and Next.js frontend.

---

🎯 Features

• 🔐 Dual-Panel Login – Admin (username/password) + User (phone/OTP)
• 📸 Real-Time Recognition – Live webcam face detection and identification
• 👑 Admin Dashboard – User management, statistics, face registration
• 🎨 Modern UI – Glassmorphism design, smooth animations, responsive
• ⚡ Fast Matching – 128-dimensional face embeddings with confidence scoring
• 📊 Recognition Logs – Track all recognition attempts

---

🛠️ Tech Stack

Backend	Frontend
FastAPI	Next.js 15
Python 3.11	TypeScript
face_recognition	TailwindCSS
SQLAlchemy	Framer Motion
SQLite	Axios
JWT Authentication	React Webcam

---

🚀 Quick Start

Prerequisites

• Python 3.11
• Node.js 18+

Backend Setup

# Clone repository
git clone https://github.com/atuljha-tech/Face-recognition-model.git
cd Face-recognition-model

# Create virtual environment
python3.11 -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Create .env file
echo "SECRET_KEY=$(python -c 'import secrets; print(secrets.token_urlsafe(32))')" > .env
echo "RATE_LIMIT_PER_MINUTE=10" >> .env

# Run server
uvicorn app.main:app --reload --port 8000
API available at: http://localhost:8000/docs

Frontend Setup
bash
cd frontend

# Install dependencies
npm install

# Create .env.local
echo "NEXT_PUBLIC_API_URL=http://localhost:8000" > .env.local

# Run development server
npm run dev
App available at: http://localhost:3000

Create Admin User
bash
curl -X POST http://localhost:8000/auth/admin/register \
  -H "Content-Type: application/json" \
  -d '{"username":"admin","email":"admin@example.com","password":"admin123"}'

🔐 Authentication-

Admin Login Username: admin (after registration) Password: admin123

User Login (Demo Mode) Any phone number works (e.g., 9876543210)

OTP: 123456 (any 6-digit OTP works)

User account auto-created on first login

🚀 Deployment-

Backend (Render) Push code to GitHub

Create new Web Service on render.com

Set: Build Command: ./render-build.sh Start Command: venv/bin/uvicorn app.main:app --host 0.0.0.0 --port $PORT

Environment: PYTHON_VERSION=3.11.9, SECRET_KEY

Frontend (Vercel) Import repository on vercel.com

Set Root Directory: frontend

Add Environment Variable: NEXT_PUBLIC_API_URL = your Render URL

📁 Project Structure

text
face-recognition-system/
├── app/                    # FastAPI backend
│   ├── models/            # Database models
│   ├── routers/           # API endpoints
│   ├── services/          # Business logic
│   └── middleware/        # Logging & rate limiting
├── frontend/              # Next.js frontend
│   ├── app/               # Pages (login, dashboard, etc.)
│   ├── components/        # Reusable UI components
│   └── utils/             # API client
├── requirements.txt       # Python dependencies
├── render-build.sh        # Render build script
└── .python-version        # Python version (3.11)

📸 How It Works Face Detection – face_recognition locates faces in images

Face Encoding – Converts face to 128-dimensional vector

Similarity Search – Compares with stored encodings

Confidence Score – Converts distance to confidence percentage

Data Augmentation – Creates 4-16 encodings per image for better accuracy

🤝 Contributing Contributions welcome! Please open an issue or submit a pull request.

📄 License MIT © Atul Jha

Made with ❤️ by Atul Jha

```