A couple of smarties at Carnegie Mellon University have come up with a new way to remove unwanted objects from digital photos. Their tool searchs Flickr and and other sites for images that match the original in composition, lighting and more. Come up with enough matches, and you can quickly replace an unsightly tourist bus or rolling billboard with info from the similar images.
The algorithm for now relies on a database of 2.3 million Flickr images, but the sample base is always expanding. Sounds like this will work best on images of oft-snapped landmarks, so don't expect instant fixes for those photos documenting your visit to the California Museum of Dredging.
.png){kind=logo}
