Author: Bagavati Narayanan
A modern take on the virtual photobooth — with artistic flair! Capture or upload photos and apply stunning artistic styles using Neural Style Transfer techniques.
Virtual Photobooth with a Twist allows users to turn ordinary photos into works of art using neural style transfer.
The app combines real-time style application, a clean frontend, and pretrained models for fast processing.
Key Highlights:
- Capture or upload images directly from the browser
- Apply artistic styles in real time
- Style inspiration via Pinterest mood boards
- Responsive and intuitive UI built with Vue.js
- Take or upload photos directly from your browser
- Apply artistic styles using Fast Neural Style Transfer
- Built-in sample style (more styles coming soon!)
- Style inspiration via Pinterest mood boards
- Clean, responsive UI built with Vue.js
| Method | Description |
|---|---|
| Neural Style Transfer (NST) | Uses a training loop to directly optimize an output image to match style and content. Highly flexible. |
| Fast Neural Style Transfer | Fast and efficient — each style requires a separate pre-trained model. Ideal for real-time applications. |
The current app uses Fast Neural Style Transfer, trained in PyTorch on Google Colab.
| Layer | Technology |
|---|---|
| Frontend | Vue.js |
| Backend API | Flask (Python) |
| Model Training | PyTorch on Google Colab |
| Style Model | Pre-trained Fast Style Transfer Model |
# Clone the repository
git clone https://github.com/your-username/virtual-photobooth.git
cd virtual-photobooth
# Backend setup
cd backend
python -m venv venv
source venv/bin/activate # Linux/macOS
venv\Scripts\activate # Windows
pip install -r requirements.txt
# Start backend server
python app.py
# Frontend setup
cd virtual-photobooth/frontend
# Install npm dependencies
npm install
# Start frontend dev server
npm run serve