search

universome / stylegan-v

[CVPR 2022] StyleGAN-V: A Continuous Video Generator with the Price, Image Quality and Perks of StyleGAN2
https://universome.github.io/stylegan-v
333 stars 35 forks source link

Face Forensics dataset preprocessing #27

Open deschanel11 opened 1 year ago

deschanel11 commented 1 year ago

Hi! I have a question about the face forensics dataset preprocessing!

Do you guys have some particular strategy for the preprocessing about the ratio that face takes from whole frame, or kind of strategy to choose videos which mostly have frontal faces only? Because I have processed my face forensic datas but still doesn't fit the data condition of yours!

Thank you!!

universome commented 1 year ago

Hi @deschanel11 , I am sorry for replying so late. We used exactly the same crops as Face Forensics provided in their dataset. We extracted the crops using this script: src/scripts/preprocess_ffs.py. The example result can be found here. Could you please elaborate on what exactly the differences are?