hi kay
thanks for your answer,
OK, in the moment i have access to my own grayscale image with
contours and shows the segmented objects in grayscale.
the problem is not the kinect, i'm looking under opencv for
seqmentation of grayscale objects, if this done, i have (i hope so) a
depth-object (body or hand)
then it should possible to create a skeleton of this object (depth ==
z directions) x,ypos are the middlepoint (contours) of this object ->
and create with opengl a 3d-object
about the IR-image ?? some plans ???
excuse my bad english
regards steven