MIT Algorithm Revolutionizes Image Segmentation, Finds Edges 50,000x More Accurately
photoxels
Courtesy of Jason Chang
Our eyes have no difficulty differentiating the edges of objects we see: e.g. a car or people walking by from the surrounding street. But computers make lots of guesses before they can do so. Now, two MIT researchers (Jason Chang, a graduate student in the Department of Electrical Engineering and Computer Science, and John Fisher, a principal research scientist at MIT’s Computer Science and Artificial Intelligence Lab) have figured out how to make this recognition process at least 50,000 times more efficient.
Did they talk to the folks at Adobe? And here I thought that Adobe had practically accomplished this feat with their new Content-Aware Fill feature and easy complex selections in Photoshop CS5.
