Learned Watershed: End-to-End Learning of Seeded Segmentation

Steffen Wolf, Lukas Schott, Ullrich Kothe, Fred Hamprecht

ICCV 2017

doi:10.1109/ICCV.2017.222 /iccv/2017/wolf2017iccv-learned/

Abstract

Learned boundary maps are known to outperform hand-crafted ones as a basis for the watershed algorithm. We show, for the first time, how to train watershed computation jointly with boundary map prediction. The estimator for the merging priorities is cast as a neural network that is convolutional (over space) and recurrent (over iterations). The latter allows learning of complex shape priors. The method gives the best known seeded segmentation results on the CREMI segmentation challenge.

PDF ICCV Semantic Scholar

Cite

Text

Wolf et al. "Learned Watershed: End-to-End Learning of Seeded Segmentation." International Conference on Computer Vision, 2017. doi:10.1109/ICCV.2017.222

Markdown

[Wolf et al. "Learned Watershed: End-to-End Learning of Seeded Segmentation." International Conference on Computer Vision, 2017.](https://mlanthology.org/iccv/2017/wolf2017iccv-learned/) doi:10.1109/ICCV.2017.222

BibTeX

@inproceedings{wolf2017iccv-learned,
  title     = {{Learned Watershed: End-to-End Learning of Seeded Segmentation}},
  author    = {Wolf, Steffen and Schott, Lukas and Kothe, Ullrich and Hamprecht, Fred},
  booktitle = {International Conference on Computer Vision},
  year      = {2017},
  doi       = {10.1109/ICCV.2017.222},
  url       = {https://mlanthology.org/iccv/2017/wolf2017iccv-learned/}
}