DAP: A Dynamic Adversarial Patch for Evading Person Detectors
Amira Guesmi, Ruitian Ding, Muhammad Abdullah Hanif, Ihsen Alouani,, Muhammad Shafique

TL;DR
This paper introduces a novel dynamic adversarial patch (DAP) that is naturalistic, robust to transformations, and more effective in evading person detectors in both digital and physical settings.
Contribution
The paper proposes a new approach for creating stealthy, robust adversarial patches using direct pixel modification and a creases transformation block, improving over GAN-based methods.
Findings
Achieves up to 82.28% success rate against YOLOv7 in digital tests.
Attains 65% success rate in physical-world scenarios.
Outperforms existing state-of-the-art adversarial attack methods.
Abstract
Patch-based adversarial attacks were proven to compromise the robustness and reliability of computer vision systems. However, their conspicuous and easily detectable nature challenge their practicality in real-world setting. To address this, recent work has proposed using Generative Adversarial Networks (GANs) to generate naturalistic patches that may not attract human attention. However, such approaches suffer from a limited latent space making it challenging to produce a patch that is efficient, stealthy, and robust to multiple real-world transformations. This paper introduces a novel approach that produces a Dynamic Adversarial Patch (DAP) designed to overcome these limitations. DAP maintains a naturalistic appearance while optimizing attack efficiency and robustness to real-world transformations. The approach involves redefining the optimization problem and introducing a novel…
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
TopicsAdversarial Robustness in Machine Learning · Generative Adversarial Networks and Image Synthesis · Advanced Neural Network Applications
