Robust Semantic Segmentation by Redundant Networks with a Layer-Specific Loss Contribution and Majority Vote

Andreas Bär, Marvin Klingner, Serin Varghese, Fabian Hüger, Peter Schlicht, Tim Fingscheidt

CVPRW 2020 pp. 1348-1358

doi:10.1109/CVPRW50498.2020.00174 /cvprw/2020/bar2020cvprw-robust/

Abstract

The lack of robustness shown by deep neural networks (DNNs) questions their deployment in safety-critical tasks, such as autonomous driving. We pick up the recently introduced redundant teacher-student frameworks (3 DNNs) and propose in this work a novel error detection and correction scheme with application to semantic segmentation. It obtains its robustnesss by an online-adapted and therefore hard-to-attack student DNN during vehicle operation, which builds upon a novel layer-dependent inverse feature matching (IFM) loss. We conduct experiments on the Cityscapes dataset showing that this loss renders the adaptive student to be more than 20% absolute mean intersection-over-union (mIoU) better than in previous works. Moreover, the entire error correction virtually always delivers the performance of the best non-attacked network, resulting in an mIoU of about 50% even under strongest attacks (instead of 1...2%), while keeping the performance on clean data at about original level (ca. 75.7%).

CVPRW Semantic Scholar

Cite

Text

Bär et al. "Robust Semantic Segmentation by Redundant Networks with a Layer-Specific Loss Contribution and Majority Vote." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2020. doi:10.1109/CVPRW50498.2020.00174

Markdown

[Bär et al. "Robust Semantic Segmentation by Redundant Networks with a Layer-Specific Loss Contribution and Majority Vote." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2020.](https://mlanthology.org/cvprw/2020/bar2020cvprw-robust/) doi:10.1109/CVPRW50498.2020.00174

BibTeX

@inproceedings{bar2020cvprw-robust,
  title     = {{Robust Semantic Segmentation by Redundant Networks with a Layer-Specific Loss Contribution and Majority Vote}},
  author    = {Bär, Andreas and Klingner, Marvin and Varghese, Serin and Hüger, Fabian and Schlicht, Peter and Fingscheidt, Tim},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2020},
  pages     = {1348-1358},
  doi       = {10.1109/CVPRW50498.2020.00174},
  url       = {https://mlanthology.org/cvprw/2020/bar2020cvprw-robust/}
}