Planar Ultrametrics for Image Segmentation
icon We study the problem of hierarchical clustering on planar graphs. We formulate this in terms of finding the closest ultrametric to a specified set of distances and solve it using an LP relaxation that leverages minimum cost perfect matching as a subroutine to efficiently explore the space of planar partitions. We apply our algorithm to the problem of hierarchical image segmentation.

Download: pdf

Text Reference

Julian Yarkony and Charless C. Fowlkes. Planar ultrametrics for image segmentation. In Neural Information Processing Systems (NIPS). 2015.

BibTeX Reference

@inproceedings{YarkonyF_NIPS_2015,
    author = "Yarkony, Julian and Fowlkes, Charless C.",
    title = "Planar Ultrametrics for Image Segmentation",
    booktitle = "Neural Information Processing Systems (NIPS)",
    year = "2015",
    tag = "grouping"
}