The App is a PyTorch implementation of the notorious CycleGAN paper by Zhu et al. (2017) “Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”. I have used and re-trained models that can be found on the official GitHub page of the project to create an easy to use web application that transforms landscape pictures into paintings of these three old masters:
- Van Gogh
- Cezanne
- Monet
The web application was built with Flask and deployed on a DigitalOcean droplet. The project wants to demonstrate its users how Generative Adversarial Networks works in production and allowed me to dive deep into the theory of GANs.