Time-lapse Mining from Internet Photos

Ricardo Martin-Brualla1     David Gallup2     Steven M. Seitz1,2
1University of Washington     2Google Inc.

See our new paper reconstructing 3D time-lapses here!

Overview

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.

Popular Press

Interview at WIRED.

Live TV interview on BBC (article, whole segment).

PBS, Gizmodo, Quartz, Engagdet, Washington Post, The Verge, CNET

Non English: interview at La Voz, Spiegel, Le Monde

Paper

Time-lapse Mining from Internet Photos (PDF | BibTeX )
Ricardo Martin-Brualla, David Gallup & Steve M. Seitz
Proceedings of ACM SIGGRAPH 2015

Upcoming

Code and more results will be published soon.

Photo Credits

Reproduced photos under Creative Commons License Attribution 2.0 Generic from Flickr users: Aliento Más Allá, jirihnidek, mcxurxo, elka_cz, Daikrieg and Free the Image. Reproduced photos under permission from Flickr user dration, Nadav Tobias, Klaus Wißkirchen and Juan Jesús Orío.