BriefGPT.xyz
May, 2020
利用深度金字塔对应关系进行子图异常检测
Sub-Image Anomaly Detection with Deep Pyramid Correspondences
HTML
PDF
Niv Cohen, Yedid Hoshen
TL;DR
本文提出了一种新的基于多分辨率特征金字塔的异常图像分割方法，该方法利用最近邻方法与深度学习特征相结合，无需预训练，能够取得无监督异常检测和区分最好的效果。
Abstract
nearest neighbor
(kNN) methods utilizing deep pre-trained features exhibit very strong
anomaly detection
performance when applied to entire images. A limitation of kNN methods is the lack of
→