The challenges? Insuring different angles of photos can be connected to those at other angles, collecting up-to-date data (perhaps this can be done with NAVTEQ's data acquisition vehicles, should the deal play out), insuring image quality...

Related applications are also interesting: What if the database sends you some images from which to select, even before you reach your destination? Further, the navigation information itself could be picture based, with each step including a picture of a landmark.

Tanja Aitamurto writing at VentureBeat notes, correctly, that this sort of tool is best used when standing still or moving slowly. The author suggests its use on college campuses or medical centers would be a great starting point. The photos also help bread down language barriers and may be particularly helpful to those with reading and other disorders.

The up-to-date issue is key here. How often do minor changes in say street names vex navigation (a recent change to East and West Emerson Streets vs. just Emerson dogged me recently)? If the realm of pictures, what if the gas station in the photo has since closed down? Despite that possibility, I expect this will be one of the next layer of navigation enhancement across the industry.