Invariance to a broad array of image corruptions, such as warping, noise, or color shifts, is an important aspect of building robust models in computer vision. Recently, several new data augmentations have been proposed that significantly improve performance on ImageNet-C, a benchmark of such corruptions. However, there is still a lack of basic understanding on the relationship between data augmentations and test-time corruptions. To this end, we develop a feature space for image transforms, and then use a new measure in this space between augmentations and corruptions called the Minimal Sample Distance to demonstrate a strong correlation between similarity and performance. We then investigate recent data augmentations and observe a significant degradation in corruption robustness when the test-time corruptions are sampled to be perceptually dissimilar from ImageNet-C in this feature space. Our results suggest that test error can be improved by training on perceptually similar augmentations, and data augmentations may not generalize well beyond the existing benchmark. We hope our results and tools will allow for more robust progress towards improving robustness to image corruptions. We provide code at https://github.com/facebookresearch/augmentation-corruption.
翻译:大量图像腐败,如扭曲、噪音或颜色变化,是建立计算机愿景中稳健模型的一个重要方面。最近,提出了若干新的数据增强建议,大大改进了图像网-C(这种腐败的基准)的绩效。然而,对于数据增强和测试时腐败之间的关系,仍然缺乏基本了解。为此,我们开发了一个图像变换的特质空间,然后在这个空间中使用称为最小样本距离的新尺度,以显示相似性和性之间的强势关系。然后,我们调查最近的数据增强,发现当测试时的腐败在测试时的稳健度明显下降,而测试时的腐败与该特征空间的图像网-C(图像网-C)明显不同。我们的结果表明,测试错误可以通过关于概念相似的增强和数据增强的培训加以改进,而数据增强可能不会超出现有基准的范围。我们希望我们的成果和工具将允许在改善图像腐败稳健度方面取得更强有力的进展。我们在https://githbub.com/paceregregistration/agregistrationmentmentment。