Computer scientists have invented a way of creating 3-D models of landmarks and geographical locations using photos available on sites like Flickr.
The technique creates the models from millions of images, processing them on a single PC with commodity graphics hardware in less than a day.
Researchers at the University of North Carolina and ETH-Zurich used all three million images of Rome available online to reconstruct all of the city’s major landmarks. They also reconstructed the landmarks of Berlin in the same way.
UNC’s Professor Jan-Michael Frahm says the process is more than a thousand times more efficient than current commercial systems such as Microsoft PhotoSynth.
“Our technique would be the equivalent of processing a stack of photos as high as the 828-meter Dubai Towers, using a single PC, versus the next best technique, which is the equivalent of processing a stack of photos 42 meters tall – as high as the ceiling of Notre Dame – using 62 PCs,” he says.
If video is available, this can also be incorporated, shortening the processing time needed for reconstruction of the models.
Frahm said eventually the models could be embedded into applications such as Google Earth or Bing Maps, allowing users to explore cities online.
It could also be a useful element of disaster response software. For example, an aircraft could be sent to take video of the aftermath of a hurricane, and the resulting 3-D model could be used to assess damage from a remote location.
There’s a video tour of Rome, here.