论文标题
最接近的邻居异常检测
Deep Nearest Neighbor Anomaly Detection
论文作者
论文摘要
最近的邻居是一种成功且长期存在的异常检测技术。最近，通过自我监督的深度方法（例如ROTNET）实现了重大进展。但是，自我监督的特征通常是表现不佳的Imagenet预先训练的功能。在这项工作中，我们调查了最近的进度是否确实可以超过在预轨特征空间上运行的最接近的邻居方法。实验表明，简单的基于最近的基于邻居的方法在：精度，射击概括，训练时间和噪声稳健性的同时，对图像分布的假设较少，在：精确度，射击概括，训练时间和噪声稳健性中均超过了自我监督的方法。
Nearest neighbors is a successful and long-standing technique for anomaly detection. Significant progress has been recently achieved by self-supervised deep methods (e.g. RotNet). Self-supervised features however typically under-perform Imagenet pre-trained features. In this work, we investigate whether the recent progress can indeed outperform nearest-neighbor methods operating on an Imagenet pretrained feature space. The simple nearest-neighbor based-approach is experimentally shown to outperform self-supervised methods in: accuracy, few shot generalization, training time and noise robustness while making fewer assumptions on image distributions.
