Issue No. 03 - March (2006 vol. 28)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TPAMI.2006.57
Leo Grady , IEEE
Spectral graph partitioning provides a powerful approach to image segmentation. We introduce an alternate idea that finds partitions with a small isoperimetric constant, requiring solution to a linear system rather than an eigenvector problem. This approach produces the high quality segmentations of spectral methods, but with improved speed and stability.
Index Terms- Graph-theoretic methods, graphs and networks, graph algorithms, image representation, special architectures, algorithms, computer vision, applications.
E. L. Schwartz and L. Grady, "Isoperimetric Graph Partitioning for Image Segmentation," in IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 28, no. , pp. 469-475, 2006.