D. Vatolin

Improving Adversarial Transferability in No-Reference Image Quality Assessment via Consensus-Guided Distillation and Local Perceptual Attack

No-reference image quality assessment (NR-IQA) metrics are widely used to evaluate and train image and video processing algorithms, but the growing reliance on deep neural networks makes these metrics vulnerable to adversarial attacks. Among such attacks, transferable black-box methods are particularly relevant in real...