In Vision's hand detection, we can work with one hand's landmarks to classify the pose. Is it possible to detect both hands' landmarks at the same time? So that we can detect a two-hand pose?
Two-hand pose detection
I'm curious about that too, even because in the presentation video they used to track multiple hands. There is even an hand instance limitation variable, set on 2 by default so I'm starting to think this is a bug, hope we'll figure it out