Adversarial Attack and Defense for Dehazing Networks. (arXiv:2303.17255v1 [cs.CV])

The research on single image dehazing task has been widely explored. However,
as far as we know, no comprehensive study has been conducted on the robustness
of the well-trained dehazing models. Therefore, there is no evidence that the
dehazing networks can resist malicious attacks. In this paper, we focus on
designing a group of attack methods based on first order gradient to verify the
robustness of the existing dehazing algorithms. By analyzing the general goal
of image dehazing task, five …

adversarial algorithms attack attacks defense focus general malicious networks order prediction research robustness single study task truth verify