I would like to be able to map a projection onto a physical body via kinect tracking, using the depth map.  Anyone have advice for best way to do this?  Ideally, it would involve a mask which is transparent where depth < X and opaque everywhere else.  then image is visible only on pixels for which depth is sufficiently shallow (i.e. a body).  But a point by point pixel loop seems computationally expensive.  Or if there is a library which does this, that would be best.