If I'm understanding you correctly, you're trying to flatten the model into a plane and then have that plane facing the camera? If so, does the model recalculate and reflatten itself on every frame, or once flattened does this fixed image face the camera?

Depending on your aplication, you may or may not need a vertex shader vs. a different model. Any chance of getting a more in-depth explination of the effect you're trying to achieve?