Efficient Decision-based Black-box Patch Attacks on Video Recognition

TL;DR

This paper introduces a novel, query-efficient decision-based patch attack method on video recognition models, addressing the challenge of limited information and high parameter space in videos, and demonstrates its effectiveness through experiments.

Contribution

The work proposes the spatial-temporal differential evolution (STDE) framework for decision-based patch attacks on videos, a novel approach that enhances attack efficiency and effectiveness.

Findings

STDE achieves state-of-the-art attack success rates.

STDE requires fewer queries than existing methods.

Attacks are highly imperceptible and localized.

Abstract

Although Deep Neural Networks (DNNs) have demonstrated excellent performance, they are vulnerable to adversarial patches that introduce perceptible and localized perturbations to the input. Generating adversarial patches on images has received much attention, while adversarial patches on videos have not been well investigated. Further, decision-based attacks, where attackers only access the predicted hard labels by querying threat models, have not been well explored on video models either, even if they are practical in real-world video recognition scenes. The absence of such studies leads to a huge gap in the robustness assessment for video models. To bridge this gap, this work first explores decision-based patch attacks on video models. We analyze that the huge parameter space brought by videos and the minimal information returned by decision-based models both greatly increase the attack difficulty and query burden. To achieve a query-efficient attack, we propose a spatial-temporal differential evolution (STDE) framework. First, STDE introduces target videos as patch textures and only adds patches on keyframes that are adaptively selected by temporal difference. Second, STDE takes minimizing the patch area as the optimization objective and adopts spatialtemporal mutation and crossover to search for the global optimum without falling into the local optimum. Experiments show STDE has demonstrated state-of-the-art performance in terms of threat, efficiency and imperceptibility. Hence, STDE has the potential to be a powerful tool for evaluating the robustness of video recognition models.