Class-Specific, Top-Down Segmentation

Abstract

In this paper we present a novel class-based segmentation method, which is guided by a stored representation of the shape of objects within a general class (such as horse images). The approach is different from bottom-up segmentation methods that primarily use the continuity of grey-level, texture, and bounding contours. We show that the method leads to markedly improved segmentation results and can deal with significant variation in shape and varying backgrounds. We discuss the relative merits of class-specific and general image-based segmentation methods and suggest how they can be usefully combined.

Cite

Text

Borenstein and Ullman. "Class-Specific, Top-Down Segmentation." European Conference on Computer Vision, 2002. doi:10.1007/3-540-47967-8_8

Markdown

[Borenstein and Ullman. "Class-Specific, Top-Down Segmentation." European Conference on Computer Vision, 2002.](https://mlanthology.org/eccv/2002/borenstein2002eccv-class/) doi:10.1007/3-540-47967-8_8

BibTeX

@inproceedings{borenstein2002eccv-class,
  title     = {{Class-Specific, Top-Down Segmentation}},
  author    = {Borenstein, Eran and Ullman, Shimon},
  booktitle = {European Conference on Computer Vision},
  year      = {2002},
  pages     = {109-124},
  doi       = {10.1007/3-540-47967-8_8},
  url       = {https://mlanthology.org/eccv/2002/borenstein2002eccv-class/}
}