search

pbaylies / stylegan-encoder

StyleGAN Encoder - converts real images to latent space
Other
740 stars 180 forks source link

Do I need align the generated data first when I want train an encoder from the scratch? #21

Closed gsygsy96 closed 5 years ago

gsygsy96 commented 5 years ago

Hi @pbaylies , I wonder whether I need align the generated data first when I want train an new encoder?

pbaylies commented 5 years ago

Hi @mehameha998 yes, you might want to do that, if possible! There's already code in there from the original repo for aligning faces.