Open ss8319 opened 1 year ago
@williamyang1991 I believe this is called 'rescaling' where the frames of the video is cropped to get just the face for VToonification.
I prefer to not rescale the video and keep it as it is. I am processing Sign Language videos and want to keep hand gestures. Is it possible? Or was VToonify build in such a way that it can only 'handle' faces.
Can I pad the frames so it doesn't get crop? Or use a different model?
How can I modify the code in this section to remove the automatic cropping. I did like to keep the output uncropped.
Or was VToonify build in such a way that it can only 'handle' faces.
VToonify is based on face toonification model (Toonify/DualStyleGAN). So basically, it will mainly change the face region in the image.
Thank you @williamyang1991
I am working in the Google Colab provided specificially, PART II - Style Transfer with specialized VToonify-D model.
Is there a direct way to add padding to Step 5: Video Toonification. Under this section, it shows that the input video needs to be rescaled.
Hi, I am wondering how to turn off automatic cropping in Video Toonification. What function and class is doing this?
I am working on processing some videos to be toonified using the Colab Notebook - PART II - Style Transfer with specialized VToonify-D model. How can I modify the code in this section to remove the automatic cropping. I did like to keep the output uncropped.
Thank you for the amazing work you are doing here.