Learning Segmentation by Random Walks Export

@inproceedings{citeulike:498897, abstract = {We present a new view of image segmentation by pairwise similarities. We interpret the similarities as edge ows in a Markov random walk and study the eigenvalues and eigenvectors of the walk's transition matrix. This interpretation shows that spectral methods for clustering and segmentation have a probabilistic foundation. In particular, we prove that the Normalized Cut method arises naturally from our framework. Finally, the framework provides a principled method for learning the...}, author = {Meila, Marina and Shi, Jianbo}, booktitle = {NIPS}, citeulike-article-id = {498897}, citeulike-linkout-0 = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.33.2235}, keywords = {graph-partitioning, random-walk, segmentation, spectral-clustering, yrbcn}, pages = {873--879}, posted-at = {2008-06-10 19:15:41}, priority = {2}, title = {Learning Segmentation by Random Walks}, url = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.33.2235}, year = {2000} }

TY - CONF ID - citeulike:498897 L3 - citeulike-article-id:498897 N2 - We present a new view of image segmentation by pairwise similarities. We interpret the similarities as edge ows in a Markov random walk and study the eigenvalues and eigenvectors of the walk's transition matrix. This interpretation shows that spectral methods for clustering and segmentation have a probabilistic foundation. In particular, we prove that the Normalized Cut method arises naturally from our framework. Finally, the framework provides a principled method for learning the... EP - 879 TI - Learning Segmentation by Random Walks T2 - NIPS SP - 873 KW - graph-partitioning KW - random-walk KW - segmentation KW - spectral-clustering KW - yrbcn AU - Meila, Marina AU - Shi, Jianbo PY - 2000/// UR - http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.33.2235 ER -