PG-Attack: A Precision-Guided Adversarial Attack Framework Against Vision Foundation Models for Autonomous Driving

TL;DR

This paper introduces PG-Attack, a novel adversarial attack framework that effectively deceives vision foundation models in autonomous driving by combining precise perturbations and deceptive text patches, highlighting vulnerabilities in safety-critical systems.

Contribution

The paper presents a new attack framework combining PMP-Attack and DTP-Attack to enhance adversarial effectiveness against vision models used in autonomous driving.

Findings

PG-Attack successfully deceives models like GPT-4V, Qwen-VL, and imp-V1.

The framework won first place in the CVPR 2024 challenge.

Codes are publicly available for further research.

Abstract

Vision foundation models are increasingly employed in autonomous driving systems due to their advanced capabilities. However, these models are susceptible to adversarial attacks, posing significant risks to the reliability and safety of autonomous vehicles. Adversaries can exploit these vulnerabilities to manipulate the vehicle's perception of its surroundings, leading to erroneous decisions and potentially catastrophic consequences. To address this challenge, we propose a novel Precision-Guided Adversarial Attack (PG-Attack) framework that combines two techniques: Precision Mask Perturbation Attack (PMP-Attack) and Deceptive Text Patch Attack (DTP-Attack). PMP-Attack precisely targets the attack region to minimize the overall perturbation while maximizing its impact on the target object's representation in the model's feature space. DTP-Attack introduces deceptive text patches that disrupt the model's understanding of the scene, further enhancing the attack's effectiveness. Our experiments demonstrate that PG-Attack successfully deceives a variety of advanced multi-modal large models, including GPT-4V, Qwen-VL, and imp-V1. Additionally, we won First-Place in the CVPR 2024 Workshop Challenge: Black-box Adversarial Attacks on Vision Foundation Models and codes are available at https://github.com/fuhaha824/PG-Attack.