Hi, thank you for your great work.
I just wondering if we already have a well-trained stylegan2 model(without the segmentation branch) can we finetune with your architecture?
Hi thank you for your interest on our work. Yes if you just fine-tune the segmentation branch, it will be similar to another work from our lab DatasetGAN https://github.com/nv-tlabs/datasetGAN_release.
Hi, thank you for your great work. I just wondering if we already have a well-trained stylegan2 model(without the segmentation branch) can we finetune with your architecture?