What Causes Optical Flow Networks to be Vulnerable to Physical
Adversarial Attacks
Computer Vision and Pattern Recognition (CVPR), 2021
- AAML
Abstract
Recent work demonstrated the lack of robustness of optical flow networks to physical, patch-based adversarial attacks. The possibility to physically attack a basic component of automotive systems is a reason for serious concerns. In this paper, we analyze the cause of the problem and show that the lack of robustness is rooted in the classical aperture problem of optical flow estimation in combination with bad choices in the details of the network architecture. We show how these mistakes can be rectified in order to make optical flow networks robust to physical, patch-based attacks.
View on arXivComments on this paper
