Enhancing the generalization performance of deep neural networks in the real world (i.e., unseen domains) is crucial for safety-critical applications 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) unexpected new scene images. These types of images are not included in the training dataset that the baseline shows a significant performance drop, contrary to ours. Being simple but 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 over existing work. Our code is available at https://github.com/shachoi/RobustNet.

本文提出了一种新型的实例选择性白化损失函数，以改善用于未见过领域的分割网络的鲁棒性；通过去除域移位的根源性样式信息，提高神经网络对视觉场景的泛化能力，实现在城市场景分割方面的卓越表现。

RobustNet：通过实例选择性白化改进城市场景分割的域泛化