Shape Sharing for Object Segmentation

We introduce a category-independent shape prior for object segmentation. Existing shape priors assume class-specific knowledge, and thus are restricted to cases where the object class is known in advance. The main insight of our approach is that shapes are often shared between objects of different categories. To exploit this “shape sharing” phenomenon, we develop a non-parametric prior that transfers object shapes from an exemplar database to a test image based on local shape matching. The transferred shape priors are then enforced in a graphcut formulation to produce a pool of object segment hypotheses. Unlike previous multiple segmentation methods, our approach benefits from global shape cues; unlike previous top-down methods, it assumes no classspecific training and thus enhances segmentation even for unfamiliar categories. On the challenging PASCAL 2010 and Berkeley Segmentation datasets, we show it outperforms the state-of-the-art in bottom-up or category-independent segmentation.