I was just wondering if there's an easy way to set up a job queue, so if I'm not at my computer when an extraction or conversion is done it will queue up another job automatically, like training or another extraction. Thoughts?
You can batch this kind of stuff, I do it often. Generally I will open the GUI and put in the settings I want for 1 video and then hit the Generate button to get the command line options.
I'll then open up an Anaconda prompt and enter the faceswap environment and execute a command similar to below (this is on Linux so won't work in Windows, but a bit of googling around should get you something similar that will work on Windows):
Code: Select all
for VIDEO in /media/vids/* ; \
do BASENAME=${VID##*/} ; \
FILENAME=${BASENAME%.*} ; \
FACEFOLDER="/media/faces/${FILENAME}" ; \
python faceswap.py extract -i "$VIDEO" -o "$FACEFOLDER" ... ; done(the ... is where the actual extract options would go)
This will iterate through all the video files in /media/vids/ and output faces into a separate folder, named by video name in /media/faces/