Researchers at the University of Washington and Google have stitched together fascinating time-lapse films based on large collections of online photos.
We introduce an approach for synthesizing time-lapse videos of popular landmarks from large community photo collections. The approach is completely automated and leverages the vast quantity of photos available online. First, we cluster 86 million photos into landmarks and popular viewpoints. Then, we sort the photos by date and warp each photo onto a common viewpoint. Finally, we stabilize the appearance of the sequence to compensate for lighting effects and minimize flicker. Our resulting time-lapses show diverse changes in the world’s most popular sites, like glaciers shrinking, skyscrapers being constructed, and waterfalls changing course.
The engadget article explains the basic process:
Here’s how it works: first, the researchers sorted some 86 million photos by geographic location, looking for widely snapped landmarks. Next, the photos were ordered by date and warped so that all had a matching viewpoint. Lastly, each photo was color-corrected to have a similar appearance, resulting in uniform time-lapse videos.
For more on the research and a PDF of the process, check out their project website at Time Lapse Mining from Internet Photos.