Exploring Robustness of Unsupervised Domain Adaptation in Semantic Segmentation
By Jinyu Yang, Chunyuan Li, Weizhi An, Hehuan Ma, Yuzhi Guo, Yu Rong, Peilin Zhao, Junzhou Huang. 2021
35 citations recorded by the source corpus.
Recent studies imply that deep neural networks are vulnerable to adversarial examples, i.e., inputs with a slight but intentional perturbation are incorrectly classified by the network. Such vulnerability makes it risky for some security-related applications (e.g., semantic segmentation in autonomous cars) and triggers tremendous concerns on the model reliability. For the first time, we comprehensively evaluate the robustness of existing UDA methods and propose a robust UDA approach. It is rooted in two observations: i) the robustness of UDA methods in semantic segmentation remains unexplored, which poses a security concern in this field; and ii) although commonly used self-supervision (e.g., rotation and jigsaw) benefits model robustness in classification and recognition tasks, they fail to provide the critical supervision signals that are essential in semantic segmentation. These observations motivate us to propose adversarial self-supervision UDA (or ASSUDA) that maximizes the agreement between clean images and their adversarial examples by a contrastive loss in the output space. Extensive empirical studies on commonly used benchmarks demonstrate that ASSUDA is resistant to adversarial attacks.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Materials Chemistry
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Serge J. Belongie
- Vladlen Koltun
- Quoc V. Le
- Trevor Darrell
- Kilian Q. Weinberger
- Sergey Levine
- Luc Van Gool
- Boyi Li
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Jitendra Malik
- Ming-Hsuan Yang
- Vladlen Koltun
- Dacheng Tao
- Xiaogang Wang
- Trevor Darrell
- Andrea Vedaldi
In Materials Chemistry
Citations in the corpus
- Language-driven Semantic Segmentation (2022, background): “Finally, a different line of work explores cross-domain adaptation methods for semantic segmentation by using feature alignment, self-training, and information propagation strategies (Yang et al., 2021; Wang et al., 2021).”