Depth Buffer - Kinect, RealSense, Astra S?

Unity Engine
1

I have a project where I need to know the positions of people standing in front of a long (18’) videow all. Have been researching some ideas and currently thinking a depth camera in the ceiling might work the best. I could feed the image to OpenCV for blob processing to figure out x,y positions on the ground.
I have a Kinect V2 and the $25 asset store Kinect wrapper but I am not having any luck getting the depth buffer image. I get gray and nothing else, color image works fine…
I have access to a RealSense camera but also fount the AstraS which uses OpenNI2…

Hoping anyone else has done something like this and can offer some advice on the best setup.

Also did some research using a plain webcam and openCV to do bg removal and that works OK but Im worried about the camera moving or changing lighting messing up the bg difference image.

2

I found if I set the compute user map setting to “body texture” in the Kinect Manager I get a good black and white image for OpenCV. Still would like to be able to access depth buffer, but I think this should work.