Audun Mathias Øygard
Audun Mathias Øygard
I assume you're meaning that you want to put new images on the face like in the face substitution demo? If so you need to get the points of the...
Using a 3d model is a bit more complicated, and there is no possibility to do that in clmtrackr currently, so you'd have to figure out some way to do...
Just to clarify, does the face not get detected at all, or is the tracking poor when you're wearing glasses? If it's about the detection, I'm not sure I can...
Could you check if it detects your face when you take off your glasses? (You can put them back on after it's detected your face :) )
I can confirm this issue, I'll try to investigate the cause.
It looks like the issue might be related to clmtrackr triggering too many updates per second. I've added a parameter in the dev-branch to limit the amount of iterations per...
Hmm, it's a bit of a bummer to have to limit updates to 60 fps (which setting maxIterationsPerAnimFrame to 1 effectively does). I've been trying to figure out what changes...
Yes, I've been thinking about making it node-compatible, but currently it's on my (longish) TODO list. :/ If anyone wants to help out, they're welcome. The main challenge is probably...
Hi, I didn't include "disgust" and "fear" in the example just because the classification of those emotions wasn't very good.
Hi! clmtrackr is using a 2D face model, so it's unfortunately not possible to get a mapping to a 3d model :(