Data-hunger and data-imbalance are two major pitfalls in many deep learning approaches. For example, on highly optimized production lines, defective samples are hardly acquired while non-defective samples come almost for free. The defects however often seem to resemble each other, e.g., scratches on different products may only differ in a few characteristics. In this work, we introduce a framework, Defect Transfer GAN (DT-GAN), which learns to represent defect types independent of and across various background products and yet can apply defect-specific styles to generate realistic defective images. An empirical study on the MVTec AD and two additional datasets showcase DT-GAN outperforms state-of-the-art image synthesis methods w.r.t. sample fidelity and diversity in defect generation. We further demonstrate benefits for a critical downstream task in manufacturing -- defect classification. Results show that the augmented data from DT-GAN provides consistent gains even in the few samples regime and reduces the error rate up to 51% compared to both traditional and advanced data augmentation methods.
翻译:数据饥饿和数据不平衡是许多深层学习方法的两大缺陷。例如,在高度优化的生产线上,很少获得有缺陷的样品,而非缺陷的样品则几乎免费获取。尽管这些缺陷似乎往往彼此相似,例如,不同产品的刮痕可能只在几个方面有所不同。在这项工作中,我们引入了一个框架,即Deffect Transport GAN(DT-GAN),它学会代表独立于各种背景产品和跨不同背景产品的缺陷类型,但可以采用针对缺陷的风格来生成现实的缺陷图像。关于MVTec AD的实证研究以及另外两个数据集展示了DT-GAN的超常态图像合成方法(w.r.t.样例)和缺陷生成中的多样性。我们进一步展示了制造中关键的下游任务 -- -- 缺陷分类 -- -- 的好处。结果显示,DT-GAN的扩大数据提供一致的收益,即使是在少数样本制度中,并将误差率降低到51%,而传统和先进的数据增强方法都是如此。