A Streamlit-based AI image processing application that combines TensorFlow and PyTorch models to perform image colorization, super-resolution enhancement, and Daltonization (color blindness simulation). The app provides an intuitive web interface for users to upload images and view processed results side-by-side.
- 🎨 Image Colorization: Convert grayscale images to color using a TensorFlow model
- 🔍 Super-Resolution Enhancement: 4x image resolution upscaling with PyTorch-based RRDBNet architecture
- 👁️ Daltonization: Simulate color blindness correction effects
- 🖼️ Responsive UI with three-column comparison view
- ⬇️ Direct image download capability
- Clone the repository
- Install dependencies:
pip install -r requirements.txt- Ensure both model files are present in the root directory:
- color_image_checkpoint.keras (TensorFlow colorization model)
- RRDB_ESRGAN_x4.pth (PyTorch super-resolution model)
- Run the Streamlit app:
streamlit run streamlit.py- Upload an image (JPG/PNG format)
- Click "Process Image" to generate:
- Original image
- AI-enhanced colorized image
- Daltonized color correction
- Download the processed image using the provided button
Contributions welcome! Please follow these guidelines:
- Fork the repository
- Create a new feature branch
- Submit a pull request with detailed description
- Follow PEP8 style guide
- Include test cases for new features
This project is licensed under the MIT License - see LICENSE file for details