BadPart: Unified Black-box Adversarial Patch Attacks against Pixel-wise Regression Tasks

arXiv:2404.00924v1 Announce Type: new
Abstract: Pixel-wise regression tasks (e.g., monocular depth estimation (MDE) and optical flow estimation (OFE)) have been widely involved in our daily life in applications like autonomous driving, augmented reality and video composition. Although certain applications are security-critical or bear societal significance, the adversarial robustness of such models are not sufficiently studied, especially in the black-box scenario. In this work, we introduce the first unified black-box adversarial patch attack framework against pixel-wise regression tasks, aiming to identify …

abstract adversarial applications arxiv attacks augmented reality autonomous autonomous driving box cs.cv daily driving flow life mde optical optical flow pixel reality regression robustness security significance tasks type video wise