search

StelaBou / StyleMask

Authors official PyTorch implementation of the "StyleMask: Disentangling the Style Space of StyleGAN2 for Neural Face Reenactment" [FG 2023].
114 stars 9 forks source link

Target and Source image weighting. #5

Closed cihankaradogan closed 1 year ago

cihankaradogan commented 1 year ago

Hello! I'm using reenactment for frontalizing faces. Gives very good result. However, it changes face a lot. I used face similarity calculations to use most similar target image but a pool with a lot of target images needed. So, I wonder is there a way to weighting source image and target image in reenactment process? With this I can easily prevent this issue. Best Regards.

StelaBou commented 1 year ago

Hello, I am not using something to weight the source and target images. However, if you want to reenact similar images then i think the only way to do it is, as you mentioned, have a large pool of images and calculate similarity scores (identity, perceptual) to generate "similar" image pairs.