Using a similar concept that the Google search engine uses, Carnegie Mellon University (CMU) researchers are trying to add information to and remove it from digital photos, by giving a computer access to extremely large amounts of information, so it can, for example, add people into another picture or remove unsightly defects from a picture by finding appropriate substitute material from a database of 2.3 million pictures. The computer doesn't have to "know" what it's looking at, only that it isn't right based on a photograph's content, geometry, and other aspects.


[Mail to a friend]