RobustNet: Improving Domain Generalization in Urban-Scene Segmentation via Instance Selective Whitening

By Sungha Choi, Sanghun Jung, Huiwon Yun, Joanne Taery Kim, Seungryong Kim, J. Choo. Computer Vision and Pattern Recognition, 2021

400 citations recorded by the source corpus.

Enhancing the generalization capability of deep neural networks to unseen domains is crucial for safety-critical applications in the real world such as autonomous driving. To address this issue, this paper proposes a novel instance selective whitening loss to improve the robustness of the segmentation networks for unseen domains. Our approach disentangles the domain-specific style and domain-invariant content encoded in higher-order statistics (i.e., feature covariance) of the feature representations and selectively removes only the style information causing domain shift. As shown in Fig. 1, our method provides reasonable predictions for (a) low-illuminated, (b) rainy, and (c) unseen structures. These types of images are not included in the training dataset, where the baseline shows a significant performance drop, contrary to ours. Being simple yet effective, our approach improves the robustness of various backbone networks without additional computational cost. We conduct extensive experiments in urban-scene segmentation and show the superiority of our approach to existing work. Our code is available at this link1.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/cd8394d8b6679bb446cf154a6d123cf9a00e561e

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

1. Fan Zhu
2. Ling Shao
3. Li Liu
4. Yi Zhou
5. Jie Qin
6. Lei Huang
7. Wanqing Li
8. Philip O. Ogunbona

In Computer Vision and Pattern Recognition

1. Li Liu
2. Ling Shao
3. Fan Zhu
4. Yi Zhou
5. Jie Qin
6. Lei Huang
7. Yi-Zhe Song
8. Jianbing Shen

Citations in the corpus

• Generalizing to Unseen Domains: A Survey on Domain Generalization (2021, neutral): “Similar spirits also include [129], [130].”
• Normalization Techniques in Training DNNs: Methodology, Analysis and Application (2020, methodology): “Furthermore, the idea of BN in learning invariant representation is used for person re-identification (ReID) [247], [248], [249] and stereo match [250], [251], where the data from different cameras can be viewed as different ‘domains’.”