mintmaker

6 fps seems about right for a 1060 MaxQ generating a Deepfake with the used model, the 47 fps are displayed when no Deepfake is generated.

@DRLTAnalysis VRAM generally helps when training something as it allows to look at more input images at the same time (you can fit more images into GPU's memory). At test...

@willfarrell On Mac it is impossible to run a Linux Docker (it is also using nvidia docker). Probably a problem when installing the linux version of pytorch. Docker is only...

First of all: avatarify just makes it easy to use the DeepFake Model [`First Order Motion Model`](https://github.com/AliaksandrSiarohin/first-order-model). So performance/image quality issues belong actually there. Anyways, I will answer your question:...

Yes, exactly but the morphing is also done by another trained neural network that probably makes the results way better. It's main task is to fill in the occlusions, which...

Okay, I thought it's a rather deep FCNN and thus one eye would be influenced by the other (similar to patch adversarial attacks influencing almost the whole output of an...

Do you mean you get a conflict when running client and server on the same machine? In that case you should add the `--is-local-client` flag (as stated in the wiki).

The first one can be safely deleted as it is not used anywhere. Not sure if the second one is right though. I will look into this as soon as...

Both can be deleted as the ports are already exposed in the Dockerfile. @alievk If you like you can just remove the whole if statement in the run.sh there. Otherwise...

@PakanAngel TL;DR No, Opencv does not support Blackmagic Capture devices and these can only be made accessible through some extra code, but it will be much effort. If you can...