Land Cover Classification From Satellite Imagery With U-Net and Lovasz-Softmax Loss

Alexander Rakhlin, Alex Davydow, Sergey Nikolenko; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2018, pp. 262-266

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 Lovasz-Softmax loss that successfully alleviates these problems; we compare several different convolutional architectures for U-Net encoders.

Related Material


[pdf]
[bibtex]
@InProceedings{Rakhlin_2018_CVPR_Workshops,
author = {Rakhlin, Alexander and Davydow, Alex and Nikolenko, Sergey},
title = {Land Cover Classification From Satellite Imagery With U-Net and Lovasz-Softmax Loss},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2018}
}