I'm currently working on a human pose estimation project, and I need to get a ground truth value out of a Kinect 2.0, I know this is possible, but I'm having trouble understanding how to do this with the freenect2 driver.
So I was wondering, is it even possible to do with the current freenect2 driver? or is there still some obscured microsoft stuff which has prevented us from gathering that data from the kinect ourselves?
I know that some of my collegues are able to just obtain 25 points corresponding to important skeletal points from projects such as this, but that depends on the kinect SDK, which I'd prefer to avoid using.