Software meshes photos to create 3D landscape

Movie Camera
  • 17:52 01 August 2006
  • NewScientist.com news service
  • Will Knight
Printable versionEmail to a friendRSS FeedSyndicate
 
Photosynth stitches overlapping pictures together to create a 3D landscape, in this case of St. Mark's Square in Venice, Italy (Image: Microsoft)
Photosynth stitches overlapping pictures together to create a 3D landscape, in this case of St. Mark's Square in Venice, Italy (Image: Microsoft)
 

Software that automatically stitches together hundreds of photographs to create a virtual world that can be explored using a computer is to be released by Microsoft.

Called Photosynth, the software takes individual images and performs careful analyses to find matching sections. It then "stitches" overlapping pictures together to create a three-dimensional landscape composed of many different snaps.

The software was developed by Noah Snavely and Steven Seitz at the University of Washington, in Seattle, US, in collaboration with Microsoft researcher Richard Szeliski.

"It's like a hybrid of a slide show and a gaming experience," says Szeliski, manager of the Interactive Visual Media Group at Microsoft Research in Washington, US. "This is a revolutionary way for people to interact with photos in a 3D context that more closely resembles the place where the images were captured."

Zoom in

Photosynth first finds distinctive features of each image. It then creates a "web" representing the relationships between different pictures. From this, the software can determine where the camera was positioned for each picture. It can then produce a "3D pointcloud" which looks as if each picture has been projected from its original camera position in three dimensions.

If a user, for example, uploads a series of holiday images featuring the same landmark they should be able explore it virtually in three dimensions and zoom in to see more features in close up.

The programmers behind the software say it could eventually be used to link up different photo collections. The research will be demonstrated at the SIGGRAPH computer graphics conference in Boston, US, on Tuesday. Microsoft has produced a video showing the software in action.

"It sounds very clever," says John Carter, an expert in computer vision at the University of Southampton in the UK, but he notes that its usefulness will only be proven over time.

Comment subject
Comment
No HTML except lower case italic tags or lower case bold tags, please:
<i> or <b>
Your name
Your email
 

We need your email in case we need to contact you about the comment. We will not use it for any other purpose.

 
 
There is 1 comment on 1 page

Great Technology! But Have A Question. . .

By Alice

Thu Feb 21 19:15:04 GMT 2008

Is it possible for me to take a picture of my dog - for example - and turn it into 3D or do we have to have 100s of them from different angles?

REPORT | REPLY

There is 1 comment on 1 page

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.

Printable versionEmail to a friendRSS FeedSyndicate
Cover of latest issue of New Scientist magazine
  • For exclusive news and expert analysis every week subscribe to New Scientist Print Edition
  • For what's in New Scientist magazine this week see contents
  • Search all stories
  • Contact us about this story
  • Sign up for our free newsletter
 
Password Login
Subscriptions