Multi-Expert Adversarial Attack Detection in Person Re-identification Using Context Inconsistency
arXiv:2108.09891
Abstract
The success of deep neural networks (DNNs) has promoted the widespread applications of person re-identification (ReID). However, ReID systems inherit the vulnerability of DNNs to malicious attacks of visually inconspicuous adversarial perturbations. Detection of adversarial attacks is, therefore, a fundamental requirement for robust ReID systems. In this work, we propose a Multi-Expert Adversarial Attack Detection (MEAAD) approach to achieve this goal by checking context inconsistency, which is suitable for any DNN-based ReID systems. Specifically, three kinds of context inconsistencies caused by adversarial attacks are employed to learn a detector for distinguishing the perturbed examples, i.e., a) the embedding distances between a perturbed query person image and its top-K retrievals are generally larger than those between a benign query image and its top-K retrievals, b) the embedding distances among the top-K retrievals of a perturbed query image are larger than those of a benign query image, c) the top-K retrievals of a benign query image obtained with multiple expert ReID models tend to be consistent, which is not preserved when attacks are present. Extensive experiments on the Market1501 and DukeMTMC-ReID datasets show that, as the first adversarial attack detection approach for ReID, MEAAD effectively detects various adversarial attacks and achieves high ROC-AUC (over 97.5%).
arXiv commentAccepted at IEEE ICCV 2021
References in corpus (13)
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- In Defense of the Triplet Loss for Person Re-Identification
- Towards Deep Learning Models Resistant to Adversarial Attacks
- Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionality
- AlignedReID: Surpassing Human-Level Performance in Person Re-Identification
- Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learning
- Distributional Smoothing with Virtual Adversarial Training
- Torchattacks: A PyTorch Repository for Adversarial Attacks
- Query Attack via Opposite-Direction Feature:Towards Robust Image Retrieval
- Motion-Excited Sampler: Video Adversarial Attack with Sparked Prior
- Connecting the Dots: Detecting Adversarial Perturbations Using Context Inconsistency
- You Do (Not) Belong Here: Detecting DPI Evasion Attacks with Context Learning
- Measurement-driven Security Analysis of Imperceptible Impersonation Attacks