Defense Against Adversarial Attacks on No-Reference Image Quality Models with Gradient Norm Regularization

arXiv:2403.11397v1 Announce Type: new
Abstract: The task of No-Reference Image Quality Assessment (NR-IQA) is to estimate the quality score of an input image without additional information. NR-IQA models play a crucial role in the media industry, aiding in performance evaluation and optimization guidance. However, these models are found to be vulnerable to adversarial attacks, which introduce imperceptible perturbations to input images, resulting in significant changes in predicted scores. In this paper, we propose a defense method to improve the stability …

abstract adversarial adversarial attacks arxiv assessment attacks cs.cv defense eess.iv evaluation gradient guidance however image industry information media norm optimization performance quality reference regularization role type