- Reviewed Version (pdf): https://openreview.net/references/pdf?id=equE_LsqFX
- Keywords: corruption robustness, data augmentation, perceptual similarity, deep learning
- Abstract: 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 there is 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 risk overfitting to the existing benchmark. We hope our results and tools will allow for more robust progress towards improving robustness to image corruptions.
- One-sentence Summary: We show that data augmentation improves error on images corrupted by transforms that are visually similar to the augmentations and that this leads to overfitting on a common corruption benchmark.
- Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics