-
[pdf]
[arXiv]
[bibtex]@InProceedings{Yu_2023_ICCV, author = {Yu, Xuanlong and Zuo, Yi and Wang, Zitao and Zhang, Xiaowen and Zhao, Jiaxuan and Yang, Yuting and Jiao, Licheng and Peng, Rui and Wang, Xinyi and Zhang, Junpei and Zhang, Kexin and Liu, Fang and Alcover-Couso, Roberto and SanMiguel, Juan C. and Escudero-Vi\~nolo, Marcos and Tian, Hanlin and Matsui, Kenta and Wang, Tianhao and Adan, Fahmy and Gao, Zhitong and He, Xuming and Bouniot, Quentin and Moghaddam, Hossein and Rai, Shyam Nandan and Cermelli, Fabio and Masone, Carlo and Pilzer, Andrea and Ricci, Elisa and Bursuc, Andrei and Solin, Arno and Trapp, Martin and Li, Rui and Yao, Angela and Chen, Wenlong and Simpson, Ivor and Campbell, Neill D. F. and Franchi, Gianni}, title = {The Robust Semantic Segmentation UNCV2023 Challenge Results}, booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops}, month = {October}, year = {2023}, pages = {4618-4628} }
The Robust Semantic Segmentation UNCV2023 Challenge Results
Abstract
This paper outlines the winning solutions employed in addressing the MUAD uncertainty quantification challenge held at ICCV 2023. The challenge was centered around semantic segmentation in urban environments, with a particular focus on natural adversarial scenarios. The report presents the results of 19 submitted entries, with numerous techniques drawing inspiration from cutting-edge uncertainty quantification methodologies presented at prominent conferences in the fields of computer vision and machine learning and journals over the past few years. Within this document, the challenge is introduced, shedding light on its purpose and objectives, which primarily revolved around enhancing the robustness of semantic segmentation in urban scenes under varying natural adversarial conditions. The report then delves into the top-performing solutions. Moreover, the document aims to provide a comprehensive overview of the diverse solutions deployed by all participants. By doing so, it seeks to offer readers a deeper insight into the array of strategies that can be leveraged to effectively handle the inherent uncertainties associated with autonomous driving and semantic segmentation, especailly within urban environments.
Related Material