Oleksandr Tkachenko

SimSwap use face parsing. Citation from SimSwap preparation: _We use the face parsing from [face-parsing.PyTorch](https://github.com/zllrunning/face-parsing.PyTorch) for image postprocessing. Please download the relative file and place it in ./parsing_model/checkpoint from [this...

> > nice job!image swap works well but video swap result skip a lot of facial alignments,certainly needs an improvment. thank you for your work! > > Please see the...

> Currently, we haven't released it yet. Please check back later. Any plans to release NVC model?

pip install --ignore-installed insightface==0.2.1

I agree, CodeFormer gives quite interesting results, especially if you experiment with the --w option. Dear @xinntao , is there any possibility to return CodeFormer to GFPGAN?

> Hello. there is two problems: > > 1. The url not exist: "model_path='https://github.com/xinntao/RealESRGAN/releases/download/v0.2.1/ReaESRGAN_x2plus.pth'" > > 2. Error on check SSL certificate: > "File "/Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/urllib/request.py", line 1345, in do_open raise...

> > > Hello. there is two problems: > > > ``` > > > 1. The url not exist: "model_path='https://github.com/xinntao/RealESRGAN/releases/download/v0.2.1/ReaESRGAN_x2plus.pth'" > > > > > > 2. Error on...

The answer to your question is clearly visible at the end of the log you posted. It's because you have CodeFormer installed in your environment, or to be more specific,...

> Hi, yesterday I have managed to successfully run both image and video tests (CPU only so far) in Colab. I will share the final notebook in GitHub tomorrow. You...

> I was under the impression that I'd be able to give a set of images for a face, and then train a model so that the face swapping results...