Friday, October 9, 2009

Computer Science Alum "Building" Rome in a Day

Computer scientists around the world are working on various projects that involve automated reconstruction of 3-D images from photos pulled off the Web. One of the researchers working in this area is Sameer Agarwal, who earned his computer science PhD here at the Jacobs School of Engineering. Agarwal is now Acting Assistant Professor of Computer Science & Engineering at the University of Washington.

National Geographic recently profiled Agarwal's "building-Rome-in-a-day" work. Read the story at

Sameer Agarwal earned his PhD in the Computer Vision Laboratory at UC San Diego, where Serge Belongie was his advisor. At UCSD, Agarwal worked on a similar project--from pictures to three dimensions--with Department of Computer Science and Engineering PhD student Manmohan Chandrakar (who just started a post-doc at UC Berkeley with Prof. Ravi Ramamoorthi). Excerpt from that 2008 story below:

Your pictures of the Grand Canyon, Times Square or other destinations may be pretty good, but wouldn’t it be nice to show them off in three dimensions? An award-winning 3D reconstruction algorithm designed by a team of computer science researchers from UC San Diego brings this dream within the grasp of reality.

This research gets at the heart of “autocalibration,” a well-studied, fundamental problem in computer vision. Autocalibration aims to recover the three dimensional structure of a scene using only its images, acquired from cameras whose internal settings and spatial orientations are unknown.

No comments: