lpk
Thanks for your reply. I download the images from the internet, maybe there are some pre-processing on them. You mean this model is not suitable for resized image ?
Sorry, I just tested the PRNet. There are some methods based on landmarks to do the face swap, maybe you can have a try.
Same problem, have you solved it ?
Hi, I also encountered this problems, could you tell me how did you solve this problems?
I also encounter with this problem, have you resolved it ?
I solved it by installing ffmpeg and add the C_INCLUDE_PATH in the /etc/profile
Same problem, maybe the initialization should be checked
Excuse me, I was confused by the weights of the losses ,have you ever tried other weights for the losses and how much influence can the weights bring to the...
@zsdonghao So, we need to comment the parameter loading lines in the main.py file?
Got it, thanks