We worked around the limitation of the model using a
GFPGANpowered a face enhancer.
The underlying library,
FaceRestoreHelper, which is used for face landscaping and dewarping, is not thread-safe and thus becomes the limiting factor.
We support Linux, MacOS and Windows.
VRAM is the bottleneck due the memory hungry
We consider 4GB as a minimum, everything above 8GB should work fine.
One thread takes around 1GB of VRAM, therefore your threads count should less or equal to that.
On Windows modify the
TEMPenvironment variable, while on Unix based operating systems, alter the
We don't allow deepfake porn due to ethical considerations.