Abstract:In this paper, we propose a novel physical stealth attack against the person detectors in real world. The proposed method generates an adversarial patch, and prints it on real clothes to make a three dimensional (3D) invisible cloak. Anyone wearing the cloak can evade the detection of person detectors and achieve stealth. We consider the impacts of those 3D physical constraints (i.e., radian, wrinkle, occlusion, angle, etc.) on person stealth attacks, and propose 3D transformations to generate 3D invisible cloak. We launch the person stealth attacks in 3D physical space instead of 2D plane by printing the adversarial patches on real clothes under challenging and complex 3D physical scenarios. The conventional and 3D transformations are performed on the patch during its optimization process. Further, we study how to generate the optimal 3D invisible cloak. Specifically, we explore how to choose input images with specific shapes and colors to generate the optimal 3D invisible cloak. Besides, after successfully making the object detector misjudge the person as other objects, we explore how to make a person completely disappeared, i.e., the person will not be detected as any objects. Finally, we present a systematic evaluation framework to methodically evaluate the performance of the proposed attack in digital domain and physical world. Experimental results in various indoor and outdoor physical scenarios show that, the proposed person stealth attack method is robust and effective even under those complex and challenging physical conditions, such as the cloak is wrinkled, obscured, curved, and from different angles. The attack success rate in digital domain (Inria data set) is 86.56%, while the static and dynamic stealth attack performance in physical world is 100% and 77%, respectively, which are significantly better than existing works.

Invisible Pedestrians: Synthesizing Adversarial Clothing Textures to Evade Industrial Camera-Based 3D Detection

Physically Realizable Natural-Looking Clothing Textures Evade Person Detectors via 3D Modeling

Adversarial Texture for Fooling Person Detectors in the Physical World

3D Invisible Cloak

Reflective Adversarial Attacks against Pedestrian Detection Systems for Vehicles at Night

Hiding from Infrared Detectors in Real World with Adversarial Clothes

Towards Effective Adversarial Textured 3D Meshes on Physical Face Recognition

Infrared Invisible Clothing:Hiding from Infrared Detectors at Multiple Angles in Real World

Physical Adversarial Attack on Vehicle Detector in the Carla Simulator

Shadow-Catcher: Looking Into Shadows to Detect Ghost Objects in Autonomous Vehicle 3D Sensing

Learning Transferable 3D Adversarial Cloaks for Deep Trained Detectors

FCA: Learning a 3D Full-Coverage Vehicle Camouflage for Multi-View Physical Adversarial Attack

Transient Adversarial 3D Projection Attacks on Object Detection in Autonomous Driving

Stealthy and Effective Physical Adversarial Attacks in Autonomous Driving

InvisibiliTee: Angle-agnostic Cloaking from Person-Tracking Systems with a Tee

Physical Adversarial Attacks Against Aerial Object Detection With Feature-Aligned Expandable Textures

Invisible Optical Adversarial Stripes on Traffic Sign against Autonomous Vehicles

Physically Adversarial Infrared Patches with Learnable Shapes and Locations

Physical 3D Adversarial Attacks against Monocular Depth Estimation in Autonomous Driving

Adversarial Obstacle Generation Against LiDAR-Based 3D Object Detection

Adversarial Point Cloud Perturbations Against 3D Object Detection in Autonomous Driving Systems