The Weizmann Institute of Science
Faculty of Mathematics and Computer Science
Computer Vision Lab
Input Image pair
User selected point-of-interest
Recovered figure-ground segmentation
Back