In case anyone is curious about seeing the system I created for my thesis in action, I have posted it for all to use.

You can try it out here.

Screenshot of NoodleNav

It requires that your web browser has the Adobe Flash player installed and that you have a webcam available for it to use (it has to track your head gestures somehow!). The main idea is that you train it by clicking the train camera button, then place your head in the box indicated (kind of a dumb training algorithm, but for now it serves its purpose), then click done. It will take a snapshot and then lock on, once you see the small box appear, hopefully around you eyes and nose area, then it will track the direction you look. This then gets translated in to panning movements on the underlying Google Maps that corresponds to the direction you turn your head. I have plans to enhance this proof of concept, but this is what I used for my thesis, so enjoy!