DeepSight

DeepSight is a full-stack media analysis application for checking images and videos with deepfake-detection models. Users can upload supported media, receive an authenticity verdict and confidence score, and optionally save scan history to an account.

Main technologies

• Frontend: React 19, Vite 8
• API server: Node.js, Express, MongoDB/Mongoose
• AI service: Python, Flask, TensorFlow, PyTorch, Hugging Face Transformers, OpenCV, NumPy
• Authentication: bcryptjs and JSON Web Tokens

Project structure

DeepSight - Final

├── ai_engine/                 # Flask AI service and media processing
│   ├── app.py
│   └── processor.py
├── client/                    # React/Vite frontend
│   ├── public/
│   ├── src/
│   ├── index.html
│   ├── package.json
│   ├── package-lock.json
│   └── vite.config.js
├── models/                    # Requires .h5 model artifacts
├── server/                    # Express API and authentication
│   ├── index.js
│   ├── package.json
│   ├── package-lock.json
│   └── test/
├── README.md
├── main.py
└── requirements.txt

Local setup

1. Install Python dependencies

python -m venv .venv
.venv\Scripts\Activate.ps1
python -m pip install -r requirements.txt

2. Install Node dependencies

cd server
npm ci
cd ..\client
npm ci
cd ..

3. Start the services

Open three terminal windows from the project root:

# Terminal 1: AI service on port 5001
python main.py

# Terminal 2: API server on port 5000
cd server
npm run dev

# Terminal 3: frontend on port 5173
cd client
npm run dev

Open the Vite URL shown in the frontend terminal, normally http://localhost:5173.

For a deployed frontend, set VITE_API_URL to the deployed API URL before building the client. The API must be able to reach the deployed AI service, and CORS_ORIGIN must include the deployed frontend URL.

Useful commands

# Build and preview the frontend
cd client
npm run build
npm run preview

# Run API tests
cd ..\server
npm test