GraphiteAI is an end-to-end web application that converts real images into high-quality, hand-drawn-style pencil sketches using a Generative Adversarial Network (GAN).
- High-Quality Sketches: Uses advanced GAN architecture (Pix2Pix/CycleGAN) for realistic transformations.
- User-Friendly Interface: Simple web interface for easy image uploads.
- Local Processing: Runs entirely on your local machine for privacy.
- Customizable: Train on your own datasets (e.g., facades, landscapes).
- Python 3.8+
pip- (Optional) GPU with CUDA support for faster inference/training
-
Clone the Repository:
git clone https://github.com/pronzzz/graphiteai.git cd graphiteai -
Install Dependencies:
pip install -r requirements.txt
Note: Ensure
flask-corsis installed. -
Download Model Weights: The application will attempt to download pre-trained weights automatically upon first run. Alternatively, you can place your own
weights.pthfile inbackend/model/.
Start the Flask server which handles the model inference:
cd backend
python app.pyThe server will start on http://localhost:5001.
Open the frontend/index.html file in your preferred web browser.
No build step is required for the frontend.
backend/: Flask API and PyTorch model definition.training/: Training scripts and dataset handling for cGAN.frontend/: Simple HTML/JS web interface.setup.sh: Script to help set up the environment (if available).GUIDE.md: Detailed guide on architecture and advanced usage.
Contributions are welcome! Please read our Contributing Guidelines for details on how to submit pull requests, report issues, and the code of conduct.
This project is licensed under the MIT License - see the LICENSE file for details.
Made with ❤️ by Pranav Dwivedi