The Cycle-Consistent Adversarial Networks (CycleGAN) is proposed by Zhu J. Y., Park T. et al. (2017) which enables machine image-to-image transformation. This transformation can generate images non-existing before from an image to another unpaired image. We aimed to implement transformation from city images in Joensuu to Monet's style images.
Cycle - GAN Structure
| Original image | Monet artistic style: |
|---|
| 
Zhu, Jun-Yan, et al. "Unpaired image-to-image translation using cycle-consistent adversarial networks." Proceedings of the IEEE international conference on computer vision. 2017.
Isola, Phillip, et al. "Image-to-image translation with conditional adversarial networks." Proceedings of the IEEE conference on computer vision and pattern recognition. 2017.