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

Over the past decade, deep learning has revolutionized conventional tasksthat rely on hand-craft feature extraction with its strong feature learningcapability, leading to substantial enhancements in traditional tasks. However,deep neural networks (DNNs) have been demonstrated to be vulnerable toadversarial examples crafted by malicious tiny noise, which is imperceptible tohuman observers but can make DNNs output the wrong result. Existing adversarialattacks can be categorized into digital and physical adversarial attacks. Theformer is designed to pursue strong attack performance in lab environmentswhile hardly remaining effective when applied to the physical world. Incontrast, the latter focus on developing physical deployable attacks, thusexhibiting more robustness in complex physical environmental conditions.Recently, with the increasing deployment of the DNN-based system in the realworld, strengthening the robustness of these systems is an emergency, whileexploring physical adversarial attacks exhaustively is the precondition. Tothis end, this paper reviews the evolution of physical adversarial attacksagainst DNN-based computer vision tasks, expecting to provide beneficialinformation for developing stronger physical adversarial attacks. Specifically,we first proposed a taxonomy to categorize the current physical adversarialattacks and grouped them. Then, we discuss the existing physical attacks andfocus on the technique for improving the robustness of physical attacks undercomplex physical environmental conditions. Finally, we discuss the issues ofthe current physical adversarial attacks to be solved and give promisingdirections.

A Survey on Physical Adversarial Attack in Computer Vision