Hi Enrico! You could also try that with some computer vision library like MediaPipe and its pose detection: https://github.com/google-ai-edge/mediapipe/blob/master/docs/solutions/pose.md
Here is a link where I posted an example on how to track hands/fingers using mediapipe and control a csound instrument with it: https://forum.csound.com/t/short-latency-sound-control/1802/3
You could theoretically build up on that by tracking pose instead of hands and, of course, make more fitting csound instrument.
Best,
Lovre