There's thousands of photos on Flickr of the same thing — landmarks, national treasures, famous places — but each photo also contains some subtle data that takes computer horsepower to use. Developers at the University of Washington used hundreds of Flickr photos to derive not just the photo, but also the photographer's position, resulting in a 3D map of the object. This is essentially what a 3D scanner does, but 3D scanners know their vantage point and the position of the data beforehand. The 3D Flickr mapping is amazing, because it uses each overlapping point as a reference to the other, and with enough of those overlapping data points it can piece them all together into a 3D model.
#
 
No comments at this time.
| 
