BriefGPT.xyz
Apr, 2023
深度图像检索中无监督多标准对抗检测
Unsupervised Multi-Criteria Adversarial Detection in Deep Image Retrieval
HTML
PDF
Yanru Xiao, Cong Wang, Xing Gao
TL;DR
本文提出了一种高效无监督的方案，通过设计三个标准以限制敌对空间，从而在深度哈希算法供应链的漏洞下识别出混淆攻击，并在实时图像查询中获得2-23%的检测率提高。
Abstract
The vulnerability in the
algorithm supply chain
of deep learning has imposed new challenges to
image retrieval systems
in the downstream. Among a variety of techniques,
→