Vulnerability of person re-identification models to metric adversarial attacks - CEA - Commissariat à l’énergie atomique et aux énergies alternatives Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

Vulnerability of person re-identification models to metric adversarial attacks

Résumé

Person re-identification (re-ID) is a key problem in smart supervision of camera networks. Over the past years, models using deep learning have become state of the art. However, it has been shown that deep neural networks are flawed with adversarial examples, i.e. human-imperceptible perturbations. Extensively studied for the task of image closedset classification, this problem can also appear in the case of open-set retrieval tasks. Indeed, recent work has shown that we can also generate adversarial examples for metric learning systems such as re-ID ones. These models remain vulnerable: when faced with adversarial examples, they fail to correctly recognize a person, which represents a security breach. These attacks are all the more dangerous as they are impossible to detect for a human operator. Attacking a metric consists in altering the distances between the feature of an attacked image and those of reference images, i.e. guides. In this article, we investigate different possible attacks depending on the number and type of guides available. From this metric attack family, two particularly effective attacks stand out. The first one, called Self Metric Attack, is a strong attack that does not need any image apart from the attacked image. The second one, called Furthest-Negative Attack, makes full use of a set of images. Attacks are evaluated on commonly used datasets: Market1501 and DukeMTMC. Finally, we propose an efficient extension of adversarial training protocol adapted to metric learning as a defense that increases the robustness of re-ID models.
Fichier principal
Vignette du fichier
41.pdf (375.7 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

cea-03251806 , version 1 (07-06-2021)

Identifiants

Citer

Quentin Bouniot, Romaric Audigier, Angelique Loesch. Vulnerability of person re-identification models to metric adversarial attacks. CVPR 2020 - IEEE/CVF Conference on Computer Visions of the and Pattern Recognition Workshops, Jun 2020, Seattle (Virtual conference), United States. pp.794-795, ⟨10.1109/CVPRW50498.2020.00405⟩. ⟨cea-03251806⟩
40 Consultations
80 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More