Radek Daněček
Radek Daněček
For this, the face detector will surely be the bottleneck. The detection from the face_alignment package that is used is not the fastest but you can definitely try to use...
if you set `save_codes` to True when running the demo, in the resulting file there will be a `posecode` vector. The first three numbers are the global rotation in axis-angle...
Sadly, this is true. As it so happens, winks like these are not very well represented in the AffectNet dataset as they don't necessarily correspond into one of the basic...
A small update, EMOCA itself and the AffectNet dataset may actually have some wink-like expressions. I have been informed, that inability to create wink shapes might run deeper than that...
This one is tricky. It might be machine specific, I am not sure. One would expect that enforcing a particular lightning version should do the trick but it seems you...
Hi, I have never used AppleARKit so I don't know what their blendshapes are exactly but if they are a standard 3DMM, I'm afraid there's no easy way to do...
It is possible. You if you have both blendshapes available, you can of course fit it to a mesh that is an output of the other. The fitting itself is...
Hello. We use exactly the same architecture for the expression encoder as DECA uses. In fact we clone the DECA encoder and then further finetune it. It's basically a standard...
Download ResNet50 from the EmotionRecognition folder and point the path to where you download it. Keep me posted on how it goes! :-) On Sun, May 1, 2022, 13:48 Filntisis...
@filby89 how did it go? :-)