用风格化合成数据提升模型抗图像退化能力
Stylized Synthetic Augmentation further improves Corruption Robustness
- 将合成图像与神经风格迁移结合进行数据增强
- 在多个基准上达到93.54%~50.86%的鲁棒准确率
- 适合关注模型在恶劣条件下的泛化性能的研究者
本文提出一种训练数据增强流程，将合成图像与神经风格迁移相结合，以缓解深度视觉模型对常见图像退化问题的脆弱性。尽管风格迁移会降低合成图像在弗雷歇起始距离（FID）指标下的质量，但这些图像在模型训练中表现出意外的益处。我们系统地分析了两种增强方法及其关键超参数对图像分类器性能的影响。结果表明，风格化与合成数据具有良好的互补性，可与TrivialAugment等主流规则型数据增强技术协同使用，但与其他方法不兼容。该方法在多个小规模图像分类基准上实现了当前最优的抗退化性能，在CIFAR-10-C、CIFAR-100-C和TinyImageNet-C上的鲁棒准确率分别达到93.54%、74.9%和50.86%。
原文摘要 · Abstract (English)
This paper proposes a training data augmentation pipeline that combines synthetic image data with neural style transfer in order to address the vulnerability of deep vision models to common corruptions. We show that although applying style transfer on synthetic images degrades their quality with respect to the common Frechet Inception Distance (FID) metric, these images are surprisingly beneficial for model training. We conduct a systematic empirical analysis of the effects of both augmentations and their key hyperparameters on the performance of image classifiers. Our results demonstrate that stylization and synthetic data complement each other well and can be combined with popular rule-based data augmentation techniques such as TrivialAugment, while not working with others. Our method achieves state-of-the-art corruption robustness on several small-scale image classification benchmarks, reaching 93.54%, 74.9% and 50.86% robust accuracy on CIFAR-10-C, CIFAR-100-C and TinyImageNet-C, respectively
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。