English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The security of object detection systems has attracted increasing attention,especially when facing adversarial patch attacks. Since patch attacks changethe pixels in a restricted area on objects, they are easy to implement in thephysical world, especially for attacking human detection systems. The existingdefenses against patch attacks are mostly applied for image classificationproblems and have difficulty resisting human detection attacks. Towards thiscritical issue, we propose an efficient and effective plug-in defense componenton the YOLO detection system, which we name Ad-YOLO. The main idea is to add apatch class on the YOLO architecture, which has a negligible inferenceincrement. Thus, Ad-YOLO is expected to directly detect both the objects ofinterest and adversarial patches. To the best of our knowledge, our approach isthe first defense strategy against human detection attacks.  We investigate Ad-YOLO's performance on the YOLOv2 baseline. To improve theability of Ad-YOLO to detect variety patches, we first use an adversarialtraining process to develop a patch dataset based on the Inria dataset, whichwe name Inria-Patch. Then, we train Ad-YOLO by a combination of Pascal VOC,Inria, and Inria-Patch datasets. With a slight drop of $0.70\%$ mAP on VOC 2007test set, Ad-YOLO achieves $80.31\%$ AP of persons, which highly outperforms$33.93\%$ AP for YOLOv2 when facing white-box patch attacks. Furthermore,compared with YOLOv2, the results facing a physical-world attack are alsoincluded to demonstrate Ad-YOLO's excellent generalization ability.

Adversarial YOLO: Defense Human Detection Patch Attacks via Detecting Adversarial Patches