Edit face in Eyeglasses, Gender, Hair Color, Pose , Smile ... direction failed

genforce / sefa

[CVPR 2021] Closed-Form Factorization of Latent Semantics in GANs

https://genforce.github.io/sefa/

MIT License

963 stars 108 forks source link

Edit face in Eyeglasses, Gender, Hair Color, Pose , Smile ... direction failed #16

Open onefish51 opened 3 years ago

onefish51 commented 3 years ago

I edited face in Eyeglasses, Gender, Hair Color, Pose , Smile ... direction by MODEL_ZOO/StyleGAN2/ffhq-1024x1024

it shown that some attributes/directions are highly correlated with each other .

MHVali commented 9 months ago

@onefish51 Hi. May I ask how did you manage to find the correct semantic indices and layer indices to manipulate for a specific direction? For example I could not find the specific semantic and layer indices for the "Smile" direction. Did the authors revealed this info somewhere? Thanks! Actually I would like to reproduce this paper's results for directions found on StyleGAN2 model trained on FFHQ dataset.