are models in faceswap just architecture or pre-trained
Posted: Tue Aug 16, 2022 8:27 pm
out of curiosity, for the models from faceswap (iae, villain, phaze-a...)
- are they just architecture, meaning we train them with our data from scratch (starting from random weights)?
or - are they pre-trained models; we are just fine-tunning them with our data (videos)?
Thanks!