Hello.
In our working setup we don't have an empty space to acquire the motion, so still cameras for us are not a satysfying solution (because of the several occlusions).
A good solution could be to follow the actor with a kinect, but it is a bit hard to do because of the lack of an embedded display (and it is a bit hard to follow the actor with kinect on one hand and a laptop on the other).
So we were considering the possibility to use a 3D camera (i.e. JVC GS-TD1, but it is just one of the several 3D cameras on the market).
What do you think about the possibility to implement an algorithm for the movement reconstruction starting from 3D cameras data? Is it possible? Is it hard to do or not?

Thanks
Rosemma