Wish you were here? Bryan Russell at the University of Washington in Seattle and colleagues have developed the first online system that automatically creates annotated 3D reconstructions of tourist sites on Wikipedia.
First, software uses Flickr images to generate a 3D model of the site, recreating its overall geometry.
Then, Wikipedia entries for the same place are scanned for nouns, which are entered into a Google image search to identify objects of interest such as paintings or statues. The photos are then added to the appropriate locations on the model. As the user scrolls through the Wikipedia article, each time a feature is mentioned, a line is drawn from the keyword to the corresponding area.
Users can zoom in on a feature or, where a narrative sequentially describes a site, they can embark on a fly-through to visualise the description.
Augmented reality
Although the labels retrieved are usually correct, the system is limited by search technology and how well-referenced a site is online. "We do quite well for the Sistine Chapel, but do worse for other sites that are not as well documented," says Russell.
Detailed analyses of the text could help improve the system. For example, identifying words that denote position, like whether an object is "above" or "to the left" of another feature, would home in on an object's location.
The team's end goal is to create a rich augmented reality application that could be used while visiting a tourist site. "By simply looking at an object, a wearable camera could recover its viewpoint relative to the 3D model, recognise the object and provide more information," says Russell. The research will be presented at Siggraph Asia in Hong Kong in November.
Fabrice Florin, product manager at the Wikimedia Foundation, which runs Wikipedia, says the site has no immediate plans to include this 3D visualization approach, but will monitor its progress. "We are excited that researchers are developing new ways to present the content of Wikipedia and other Wikimedia projects," he says.
If you would like to reuse any content from New Scientist, either in print or online, please contact the syndication department first for permission. New Scientist does not own rights to photos, but there are a variety of licensing options available for use of articles and graphics we own the copyright to.
Have your say
Only subscribers may leave comments on this article. Please log in.
Only personal subscribers may leave comments on this article
All comments should respect the New Scientist House Rules. If you think a particular comment breaks these rules then please use the "Report" link in that comment to report it to us.
If you are having a technical problem posting a comment, please contact technical support.