Land Cover Classification from Satellite Imagery with U-Net and Lovasz-SoftMax Loss
Abstract
The land cover classification task of the DeepGlobe Challenge presents significant obstacles even to state of the art segmentation models due to a small amount of data, incomplete and sometimes incorrect labeling, and highly imbalanced classes. In this work, we show an approach based on the U-Net architecture with the Lov´asz-Softmax loss that successfully alleviates these problems; we compare several different convolutional architectures for U-Net encoders.
Cite
Text
Rakhlin et al. "Land Cover Classification from Satellite Imagery with U-Net and Lovasz-SoftMax Loss." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2018. doi:10.1109/CVPRW.2018.00048Markdown
[Rakhlin et al. "Land Cover Classification from Satellite Imagery with U-Net and Lovasz-SoftMax Loss." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2018.](https://mlanthology.org/cvprw/2018/rakhlin2018cvprw-land/) doi:10.1109/CVPRW.2018.00048BibTeX
@inproceedings{rakhlin2018cvprw-land,
title = {{Land Cover Classification from Satellite Imagery with U-Net and Lovasz-SoftMax Loss}},
author = {Rakhlin, Alexander and Davydow, Alex and Nikolenko, Sergey I.},
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
year = {2018},
pages = {262-266},
doi = {10.1109/CVPRW.2018.00048},
url = {https://mlanthology.org/cvprw/2018/rakhlin2018cvprw-land/}
}