Recent advances in person re-identification (re-ID) have led to impressive retrieval accuracy. However, existing re-ID models are challenged by the adversarial examples crafted by adding quasi-imperceptible perturbations. Moreover, re-ID systems face the domain shift issue that training and testing domains are not consistent. In this study, we argue that learning powerful attackers with high universality that works well on unseen domains is an important step in promoting the robustness of re-ID systems. Therefore, we introduce a novel universal attack algorithm called "MetaAttack" for person re-ID. MetaAttack can mislead re-ID models on unseen domains by a universal adversarial perturbation. Specifically, to capture common patterns across different domains, we propose a meta-learning scheme to seek the universal perturbation via the gradient interaction between meta-train and meta-test formed by two datasets. We also take advantage of a virtual dataset (PersonX), instead of real ones, to conduct meta-test. This scheme not only enables us to learn with more comprehensive variation factors but also mitigates the negative effects caused by biased factors of real datasets. Experiments on three large-scale re-ID datasets demonstrate the effectiveness of our method in attacking re-ID models on unseen domains. Our final visualization results reveal some new properties of existing re-ID systems, which can guide us in designing a more robust reID model. Code and supplemental material are available at https://github.com/FlyingRoastDuck/MetaAttack_AAAI21.

Learning to Attack Real-World Models for Person Re-identification via Virtual-Guided Meta-Learning / Yang, Fengxiang; Zhong, Zhun; Liu, Hong; Wang, Zheng; Luo, Zhiming; Li, Shaozi; Sebe, Nicu; Satoh, Shin'Ichi. - (2021), pp. 3128-3135. (Intervento presentato al convegno The Thirty-Fifth AAAI Conference on Artificial Intelligence, The Thirty-Third Conference on Innovative Applications of Artificial Intelligence, The Eleventh Symposium on Educational Advances in Artificial Intelligence tenutosi a online nel 2nd–9th February 2021).

Learning to Attack Real-World Models for Person Re-identification via Virtual-Guided Meta-Learning

Zhong, Zhun;Sebe, Nicu;
2021-01-01

Abstract

Recent advances in person re-identification (re-ID) have led to impressive retrieval accuracy. However, existing re-ID models are challenged by the adversarial examples crafted by adding quasi-imperceptible perturbations. Moreover, re-ID systems face the domain shift issue that training and testing domains are not consistent. In this study, we argue that learning powerful attackers with high universality that works well on unseen domains is an important step in promoting the robustness of re-ID systems. Therefore, we introduce a novel universal attack algorithm called "MetaAttack" for person re-ID. MetaAttack can mislead re-ID models on unseen domains by a universal adversarial perturbation. Specifically, to capture common patterns across different domains, we propose a meta-learning scheme to seek the universal perturbation via the gradient interaction between meta-train and meta-test formed by two datasets. We also take advantage of a virtual dataset (PersonX), instead of real ones, to conduct meta-test. This scheme not only enables us to learn with more comprehensive variation factors but also mitigates the negative effects caused by biased factors of real datasets. Experiments on three large-scale re-ID datasets demonstrate the effectiveness of our method in attacking re-ID models on unseen domains. Our final visualization results reveal some new properties of existing re-ID systems, which can guide us in designing a more robust reID model. Code and supplemental material are available at https://github.com/FlyingRoastDuck/MetaAttack_AAAI21.
2021
AAAI-21, IAAI-2, and EAAI-21 Proceedings
Palo Alto, California, USA
AAAI
978-1-57735-866-4
Yang, Fengxiang; Zhong, Zhun; Liu, Hong; Wang, Zheng; Luo, Zhiming; Li, Shaozi; Sebe, Nicu; Satoh, Shin'Ichi
Learning to Attack Real-World Models for Person Re-identification via Virtual-Guided Meta-Learning / Yang, Fengxiang; Zhong, Zhun; Liu, Hong; Wang, Zheng; Luo, Zhiming; Li, Shaozi; Sebe, Nicu; Satoh, Shin'Ichi. - (2021), pp. 3128-3135. (Intervento presentato al convegno The Thirty-Fifth AAAI Conference on Artificial Intelligence, The Thirty-Third Conference on Innovative Applications of Artificial Intelligence, The Eleventh Symposium on Educational Advances in Artificial Intelligence tenutosi a online nel 2nd–9th February 2021).
File in questo prodotto:
File Dimensione Formato  
16422-Article Text-19916-1-2-20210518-1.pdf

Solo gestori archivio

Tipologia: Versione editoriale (Publisher’s layout)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 4.25 MB
Formato Adobe PDF
4.25 MB Adobe PDF   Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11572/326184
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 10
  • ???jsp.display-item.citation.isi??? 4
social impact