The Digital Insider

Along similar lines as yesterday’s post, I came across another project from a team at Carnegie Mellon. The team has created a project called fotowoosh which works to create a 3D image out of a 2 dimensional digital photography. The program attempts to recreate the way that humans extract information from a 2 dimensional photo, using our knowledge of the physical world to interpret the 3 dimensional nature of the photographed scene. We instinctually use our knowledge of the world to create a 3 dimensional view in our mind.

The team’s approach is to write a program which learns the structure of the world and the appearance of geometric surfaces from a large set of training images. The knowledge is then applied to other photos. If the program can determine where the vertical surfaces contact the ground in the image, the depth of generating a 3D model.

The video below demonstrates the power of this new program.

This entry was posted on Tuesday, August 21st, 2007 at 7:17 pm and is filed under Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.