Automatically infuse grayscale images with vivid colors using CycleGAN, a powerful Generative Adversarial Network (GAN) architecture. This project leverages Python, TensorFlow, and Keras to achieve high-quality image colorization and seamless conversion between grayscale and color versions.
CycleGAN is a type of GAN that can learn to translate images from one domain to another without the need for paired training data. In this project, we train the CycleGAN model on a dataset of grayscale images and their corresponding color versions. Once trained, the model can take any grayscale image as input and produce a realistic colorized version as output.
- Clone this repository to your local machine.
- Install the required libraries and dependencies using
pip install -r requirements.txt
.
- Ensure you have the grayscale images you want to colorize in a separate folder.
- Use
Inference.ipynb
to Convert them
Special thanks to the developers of CycleGAN and the TensorFlow community for their invaluable contributions.
Enjoy transforming your grayscale images into vibrant, colorful masterpieces!