Sorry about the information i missed in my last post. I have already transformed the camera data to world coordinates considering the tilt and other external parameters. But the problem is how do i calculate length, breadth or height of individual segments from their associated point cloud.
As i have a grid of points for each segment i need to know how to consider these points to calculate length, breadth or height.
For e.g., if a segment is such that it is directly facing the camera, to know its height, it is sufficient to calculate the distance between the points on the top most row and the bottom most row of the segment. The points which are in between need not be considered.
obviously this heuristic works for a special case. I would like to know if there is a more general way to address this issue.
I haven't considered kinect yet.