Photo tourism is a system for browsing large collections of photographs in 3D. Our approach takes as input large collections of images from either personal photo collections or Internet photo sharing sites (a)
, and automatically computes each photo's viewpoint and a sparse 3D model of the scene (b)
. Our photo explorer interface enables the viewer to interactively move about the 3D space by seamlessly transitioning between photographs, based on user control (c)
Visit our PhotoCity
project where you can help us reconstruct
your neck of the woods.
We have also scaled up the vision
algorithms behind Photo Tourism to
work on entire cities. See the Rome
in a Day project page for more details.
See our work
Paths through the World's
Our structure from
motion code is also now available at
Short Video - 1 min [wmv 11 MB] [mov 23 MB]
Long Video - 5.5 min [wmv 55 MB] [mov 120 MB]
Noah Snavely, Steven M. Seitz, Richard Szeliski, "Photo tourism:
Exploring photo collections in 3D," ACM Transactions on Graphics
(SIGGRAPH Proceedings), 25(3), 2006, 835-846. [pdf] [bibtex]
For more details on the structure from motion algorithm, please see:
Noah Snavely, Steven M. Seitz, Richard Szeliski, "Modeling the world from Internet photo collections," International Journal of Computer Vision (to be published). [pdf] [bibtex]
For information about how to
extract focal length data from EXIF tags, please visit this page.
For information on other related projects, please visit the UW Community Photo Collections project page.
- Full Presentation - [zip 164 MB]
- Presentation without videos - [ppt 35 MB]
and Reconstructions - We
are making several Internet image
collections (and associated
reconstructions) available here.
Database - Simon Winder and Matt Brown have
used our data to create a database of corresponding local image
patches, available here.