Ivan Skorokhodov

There are three possible choices for augmentations: 1) no augmentations (`aug: noaug`), 2) fixed augmentations (`aug: fixed`); and 3) adaptive augmentations (`aug: ada`). If you choose to have adaptive augmentations,...

Hi, Lee! Could you please tell how exactly you change the lines `24` and `25`? Do you change them by removing the `src` prefix? I.e., from: ``` import src.legacy from...

Hi! Sorry for such a late reply. I have just pushed our implementation on top of NVidia's [StyleGAN2-ADA repo](https://github.com/NVlabs/stylegan2-ada-pytorch). It is not in pytorch-lightning (since NVidia's repo is not), but...

We wanted to show that for our model, the geometry information is stored in a very "disentangled" manner in the latent code. If your keypoints estimator is very powerful (in...

Hi, thanks! I suppose that this pixelation issue is mainly due to the spatial independence between pixels. The spatial independence between pixels makes it difficult for the model to make...

Hi, thank you! Could you please tell whether you use the FFHQ dataset with pre-computed camera poses or without them? The FFHQ dataset which we use is located in [`configs/dataset/ffhq_posed.yaml`](https://github.com/universome/epigraf/blob/main/configs/dataset/ffhq_posed.yaml)

Could you please tell in what way does the model fail for you? Does it fail to learn the texture or structure or both? If that's possible, can you please...

Hi! Turned out the problem was during printing G's summary. Looks like even using `torch.no_grad()` context does not help — likely due to some hooks inside the `print_module_summary(...)` function. I...

Hi Mohit! Yes, we plan to release the models, but we wanted to train them from scratch in the current repo to be completely sure that this repo after the...

Hi, for FFHQ, you can download the camera poses from [GRAM](https://github.com/microsoft/GRAM). For Cats, it is also available in the same repo.