ML Anthology

Boundary Extraction in Natural Images Using Ultrametric Contour Maps

Pablo Arbeláez
CVPRW 2006 pp. 182
doi:10.1109/CVPRW.2006.48 /cvprw/2006/arbelaez2006cvprw-boundary/

Abstract

This paper presents a low-level system for boundary extraction and segmentation of natural images and the evaluation of its performance. We study the problem in the
\nframework of hierarchical classification, where the geometric structure of an image can be represented by an ultrametric contour map, the soft boundary image associated to
\na family of nested segmentations. We define generic ultrametric distances by integrating local contour cues along the
\nregions boundaries and combining this information with region attributes. Then, we evaluate quantitatively our results with respect to ground-truth segmentation data, proving that our system outperforms significantly two widely
\nused hierarchical segmentation techniques, as well as the
\nstate of the art in local edge detection.

PDF CVPRW Semantic Scholar

Cite

Text

Arbeláez. "Boundary Extraction in Natural Images Using Ultrametric Contour Maps." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2006. doi:10.1109/CVPRW.2006.48

Markdown

[Arbeláez. "Boundary Extraction in Natural Images Using Ultrametric Contour Maps." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2006.](https://mlanthology.org/cvprw/2006/arbelaez2006cvprw-boundary/) doi:10.1109/CVPRW.2006.48

BibTeX

@inproceedings{arbelaez2006cvprw-boundary,
  title     = {{Boundary Extraction in Natural Images Using Ultrametric Contour Maps}},
  author    = {Arbeláez, Pablo},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2006},
  pages     = {182},
  doi       = {10.1109/CVPRW.2006.48},
  url       = {https://mlanthology.org/cvprw/2006/arbelaez2006cvprw-boundary/}
}