Abstract
We propose an incremental improvement to Fully Convolutional Data Description (FCDD), an adaptation of the one-class classification approach from anomaly detection to image anomaly segmentation (a.k.a. anomaly localization). We analyze its original loss function and propose a substitute that better resembles its predecessor, the Hypersphere Classifier (HSC). Both are compared on the MVTec Anomaly Detection Dataset (MVTec-AD) -- training images are flawless objects/textures and the goal is to segment unseen defects -- showing that consistent improvement is achieved by better designing the pixel-wise supervision.
Abstract (translated)
我们提议对全卷积数据描述(FCDD)进行增量改进,它是一类分类方法从异常检测到图像异常分割(也称为异常定位)的适应方法。我们对原损失函数进行分析并提出了一种更好的替代方案,它更类似于其前代——超平面分类器(HSC)。在MVTec异常检测数据集(MVTec-AD)上,这两种方法都进行了比较,训练图像都是完美的对象/纹理,目标是分割未看到的的缺陷。结果显示,通过更好的像素级监督设计,可以取得一致的改进。
URL
https://arxiv.org/abs/2301.09602