Invisibility Cloak: Disappearance under Human Pose Estimation via Backdoor Attacks
Minxing Zhang, Wenshu Fan, Wenbo Jiang, Shui Yu, Michael Backes, Xiao Zhang

TL;DR
This paper reveals the vulnerability of human pose estimation systems to backdoor attacks that can cause the system to fail in recognizing human poses, posing safety risks in autonomous applications.
Contribution
It introduces a novel framework, IntC, for crafting invisibility cloaks in HPE models using backdoor techniques, demonstrating effective disappearance attacks across different HPE methods.
Findings
IntC successfully causes HPE systems to ignore human poses.
The attacks are highly stealthy and generalizable across various HPE techniques.
Experiments confirm the effectiveness of the proposed backdoor methods.
Abstract
Despite being significant in autonomous systems, Human Pose Estimation (HPE)'s potential risks to adversarial attacks have not received comparable attention with image classification or segmentation tasks. In this paper, we study the vulnerability of HPE systems to disappearance attacks, where the attacker aims to subtly alter the HPE training process via backdoor techniques so that any input image with some specific trigger will not be recognized as involving any human pose. As humans are typically at the center of HPE systems, a successful attack will severely threaten pedestrians' lives if a self-driving car incorrectly understands the front scene. To achieve the adversarial goal of disappearance, we propose \emph{IntC}, a general framework to craft an invisibility cloak in the HPE domain. By designing target HPE labels that do not represent any human pose, we propose three specific…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsFace recognition and analysis · Video Surveillance and Tracking Methods · Gait Recognition and Analysis
