Andrew,
3D point tracking can be a complex problem. 3D points can not be easily generated from a 2D image - at least not accurately. You might be able to calculate the 'depth' of an object through perspective, but you would have to know the size of the object to begin with.
Normally, stereoscopy is used to accomplish this task. This technique uses two cameras separated by a known distance. By analyzing the difference between the two images, depth of field can be calculated for objects.
The following link is a paper dealing with the same issue:
http://www.ics.forth.gr/cvrl/publications/journals/2004_04_aoip_vol43_No2_3d_skin_tracking.pdf
If you are looking for a simple solution, two cameras mounted at 90 degrees from each other would allow measuring distances in the different planes with the IMAQ caliper/edge detection/ or pattern match functions from the IMAQ Vision Development Module.
I hope this gives you some ideas and possible directions. Good luck with your project.