Scale-Invariant Adversarial Attack against Arbitrary-scale Super-resolution
This addresses a security problem for users of arbitrary-scale SR models, but it is incremental as it builds on existing adversarial attack research for fixed-scale SR.
The paper tackles the vulnerability of arbitrary-scale super-resolution (SR) models to adversarial attacks by proposing a scale-invariant attack method called SIAGT, which reduces time and memory consumption while maintaining effectiveness, as demonstrated by experiments on three LIIF-based SR approaches and four datasets.
The advent of local continuous image function (LIIF) has garnered significant attention for arbitrary-scale super-resolution (SR) techniques. However, while the vulnerabilities of fixed-scale SR have been assessed, the robustness of continuous representation-based arbitrary-scale SR against adversarial attacks remains an area warranting further exploration. The elaborately designed adversarial attacks for fixed-scale SR are scale-dependent, which will cause time-consuming and memory-consuming problems when applied to arbitrary-scale SR. To address this concern, we propose a simple yet effective ``scale-invariant'' SR adversarial attack method with good transferability, termed SIAGT. Specifically, we propose to construct resource-saving attacks by exploiting finite discrete points of continuous representation. In addition, we formulate a coordinate-dependent loss to enhance the cross-model transferability of the attack. The attack can significantly deteriorate the SR images while introducing imperceptible distortion to the targeted low-resolution (LR) images. Experiments carried out on three popular LIIF-based SR approaches and four classical SR datasets show remarkable attack performance and transferability of SIAGT.