Align kinect skeleton with rgb and depth image

vvvv beta question
1

Hi,

I want to align the skeleton with the RGB and Depth streams. The openNI node does come with a switch to turn on registration between depth and rgb. the microsoft sdk does have functions to transform coordinates between the coordinate systems but as far as I have seen, there is nothing in the nodes. Has anyone done this in vvvv?

Best
Chris

2

you can take 2d points from MS SDK.
also you can use node Perspective (Transform Kinect), so when you connect it to a renderer in theory you have same perspective as kinect, so then you need to position you rgb and depth quads, or you can do a trick with z-write…

3

thanks for your reply. I cannot find a node that provides 2d skeleton points.
Perspective (Transform Kinect) connected to the renderer worked to align the joint data with the depth stream. the microsoft nodes rgb image still needs a non linear transformation to fit the depth and skeleton. whats the trick with z write?

4

hi all, looking around on the forum i’m searching the same solution. basically aligning depth map (and player) with rgb. but seems that nobody has a solution for this. i saw on the kinect node (ms) 2 interesting output pins: “color fovxy” and “depth fovxy”…
is that a good way to proceed? and how? someone has a solution?

5

continued here: https://discourse.vvvv.org/