Ivan Skorokhodov

Hi! Sorry for replying that late. We plan to release the code within the couple of weeks. If you like, here is a (very) dirty, unrefactored version of the code:...

Hi, @GuptaSonam! We are currently working on the camera-ready version of our paper (with the deadline on March 28) and will release the final version of the code around that...

Hi all! We've just pushed the majority of the code to this repo and we'll add the checkpoints/jupyter notebooks/CLIP editing scripts a bit later

Hi, thank you! It's true that FVD is not deterministic, but it is also true for FID, because one uses random samples to compute the fake statistics from — though...

Hi! Yeah, we do not report the MEAD results in the main table since we could only run StyleGAN-V and MoCoGAN-HD on it (DIGAN seems to be expensive to run...

Hi! For their repos, we didn't integrate our FVD evaluation into them, but rather sampled from the models to construct a dataset of fake videos and then used our [src/scripts/calc_metrics_for_dataset.py](https://github.com/universome/stylegan-v/blob/master/src/scripts/calc_metrics_for_dataset.py)...

Hi! To be honest, I believe that the issue you report does not seem to be easily fixable. I attribute it to the fact that the generator uses just a...

Hi! The inverted images are being generated due to the use of differentiable augmentations (from StyleGAN2-ADA). The white/black background happens for the same reason. If your generator produces the inverted...

Ok, sounds good. Also note that computing FVD on a small amount of videos (100 instead of 2000) might lead to worse FVD values because it will think that you...

Hi! The question you are asking is somewhat difficult since it is difficult to predict how the model would perform with these or that augmentations. I believe that you would...