Carnegie Mellon University researchers are applying Google's concepts for sifting through data to the tasks of adding or removing images from digital photos.
-
Carnegie Mellon University researchers have developed techniques to let computers find appropriate matches of imagery that can be added to a photo, matching lighting, perspective and other parameters. In this instance, the computer sifted through a large database of images of people and found ones that could be grafted into an image of Abbey Road, the street immortalized by The Beatles' Abbey Road album cover. Researchers call this technology Photo Clip Art.
Credit: Carnegie Mellon University -
Carnegie Mellon's Scene Completion technology can find appropriate fill-in material for those who want to trim unsightly elements out of photos. Here, a computer sifted through a database of 2.3 million Flickr images for one that could replace the building with seawater in this Mediterranean scene. The technology presents the computer user with a selection of the best 20 matches. The area of sea covered with sailboats was taken from a completely different image.
Credit: Carnegie Mellon University
Add Your Comment 1