No idea if this will be good enough, and it will face the usual problems with people being skeptical, sneering, and whether it will be good enough in terms of latency and coping with stage conditions and field of view limitations.
But I keep an eye on this stuff because I expect there may be potential to add the missing slide gesture to the likes of Osmose via this sort of approach eventually, although again I make no claims with regards this particular device. In my mind I sort of have an assumption that eventually realtime image processing will enable even suboptimal camera angles, conditions etc to be able to offer a clean enough signal to make this sort of approach even more viable, but Ive absolutely no idea how many years that might take and perhaps other types of sensor will always have the edge.
When ROLI did their Leap Motion based stuff I was hoping they might lean in that direction, but of course they had a diffeent sort of gimmick in mind and had all their usual problems as a company with huge debt etc to contend with.
Demo isnt really good enough in this particular video either, especially not the rotation aspect, so Im mostly just keeping an eye on the moving up and down a key aspect.