Recent studies have proven that DNNs, unlike human vision, tend to exploit texture information rather than shape. Such texture bias is one of the factors for the poor generalization performance of DNNs. We observe that the texture bias negatively affects not only in-domain generalization but also out-of-distribution generalization, i.e., Domain Generalization. Motivated by the observation, we propose a new framework to reduce the texture bias of a model by a novel optimization-based data augmentation, dubbed Stylized Dream. Our framework utilizes adaptive instance normalization (AdaIN) to augment the style of an original image yet preserve the content. We then adopt a regularization loss to predict consistent outputs between Stylized Dream and original images, which encourages the model to learn shape-based representations. Extensive experiments show that the proposed method achieves state-of-the-art performance in out-of-distribution settings on public benchmark datasets: PACS, VLCS, OfficeHome, TerraIncognita, and DomainNet.

翻译：近期研究证明，深度神经网络（DNN）与人类视觉不同，倾向于利用纹理信息而非形状特征。这种纹理偏差是导致DNN泛化性能较差的因素之一。我们观察到，纹理偏差不仅影响域内泛化，还会对分布外泛化（即域泛化）产生负面影响。受此观察启发，我们提出一种基于新型优化数据增强的框架——风格化梦境（Stylized Dream），以降低模型的纹理偏差。该框架利用自适应实例归一化（AdaIN）增强原始图像的风格特征同时保留内容信息。随后采用正则化损失函数促使模型对风格化梦境图像与原始图像输出一致的结果，从而引导模型学习基于形状的表征。大量实验表明，所提方法在PACS、VLCS、OfficeHome、TerraIncognita和DomainNet等公开基准数据集的分布外设置中均达到最先进性能。