Kinect offered good full body skeletal tracking. Saying that it could “only recognize dancing” is objectively wrong. And even if that was the issue, look at eg

Kinect offered good full body skeletal tracking. Saying that it could “only recognize dancing” is objectively wrong. And even if that was the issue, look at eg Magic Leap, which had excellent finger tracking. What compelling things were built for it? The issue is that “gestures” are not a binary thing that you do or not; they live on a continuum. What is waving vs raising your hand to swat a fly? Categorizing these “gestures” in a meaningful way and layering effective gameplay on top of it is the challenge, that pretty much no one solved despite the huge incentives. Gestures on two-dimensional multitouch screens are not what we’re talking about here (and even then, you can’t expect people to remember much more beyond the 5/6 core ones).