You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have another questions, if the dataset ground truth is generated by "other model" and you use these data for training, does it mean your model performance limit is determined by the ground truth generation model?
the HQ imgs seem have artifacts in face, will you select and remove the bad quality img?
The text was updated successfully, but these errors were encountered: