Is High-Resolution Image-to-Image Translation Possible - Pre-Trained + Fine-Tuning?

[jacojoubert12]

trafficstars

I would like to know if High-Resolution Image-to-Image Translation is possible using this repo by using a pre-trained model and fine-tuning it on a custom dataset. For example, BERT NLP models can be used with something like 'from-pretrained', and then a custom dataset is added for fine tuning on specific text.

I've seen other repos do this in the form of directory structures containing the images like this: TrainA TrainB TestA TestB

It is very easy to use, but this repo looks like it can provide superior results if this is possible and straightforward as above.

It is not clear to me from the README that simple image-to-image translation is possible, starting with a pretrained model and just doing fine tuning. Any advice will be greatly appreciated.