IM2GPS: estimating geographic information from a single image

Google Tech Talks
August 5, 2008

ABSTRACT

Estimating geographic information from an image is an excellent, difficult high-level computer vision problem whose time has come. The emergence of vast amounts of geographically-calibrated image data is a great reason for computer vision to start looking globally on the scale of the entire planet! In this paper, we propose a simple algorithm for estimating a distribution over geographic locations from a single image using a purely data-driven scene matching approach. For this task, we will leverage a dataset of over 6 million GPS-tagged images from the Internet. We represent the estimated image location as a probability distribution over the Earth's surface. We quantitatively evaluate our approach in several geolocation tasks and demonstrate encouraging performance (up to 30 times better than chance). We show that geolocation estimates can provide the basis for numerous other image understanding tasks such as population density estimation, land cover estimation or urban/rural classification.

Speaker: James Hays
James Hays received his B.S. in Computer Science from Georgia Institute of Technology in 2003. He has been a Ph.D. student in Carnegie Mellon University's Computer Science Department since 2003 and is advised by Alexei A. Efros. His research interests are in computer vision and computer graphics, focusing on image understanding and manipulation leveraging massive amounts of data. His research has been supported by a National Science Foundation Graduate Research Fellowship.

Category:

People & Blogs

Tags:

• google
• techtalks
• techtalk
• engedu
• talk
• talks
• googletechtalks
• education

[suhailmerchant]

The listeners had some great inputs, makes me think that was probably the reason for this lecture..

[quantumG]

This tech talk should be renamed to "doing similar image search with geotagging biasing".. cause that's all he's doing.

[mndrix]

He describes a clever technique for geolocating an image based only on its appearance. The examples get a bit repetitive.  There are some good suggestions for future improvement at the end.

[stimpy100]

Very interesting

[kissitordie]

videos rock.