Researchers Co-Op Flickr with 3-D Effects

Flickr has become the modern day slide show, but who knew the photo sharing site could help build 3-D composites of our most famous landmarks. Researchers at the University of Washington have developed a piece of software that synthesizes thousands of photos, digitally edits out your kids cheesing for the camera, calculates where each photographer was standing when they took the shot. By comparing two photos taken from slightly different perspectives, the software applies principles of computer vision to figure out the distance to each point. The app then stitches the photos together to create a highly detailed 3-D model of the Statue of Liberty, the Grand Canyon, or any number of tourist destinations getting snapped by the minute.

The challenge to using random travel snaps is there's no consistency from one shot to the next. Lighting is different. Angles vary. Image quality runs the gamut. So the researchers created algorithms to filter out extraneous data and meld the images.

The current modeling lacks color and texture. In time, researchers hope to create city models far more detailed than views from Google Earth.

[via Business Week; image via UW News]