Exploring the differences in adversarial robustness between ViT- and CNN-based models using novel metrics

• Published in: Computer vision and image understanding Vol. 235; p. 103800
• Main Authors: Heo, Jaehyuk, Seo, Seungwan, Kang, Pilsung
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.10.2023
• Subjects: Adversarial robustness; Computer vision; 65D05; Computer vision; 65D17; Adversarial robustness; 41A05; 41A10

Deep-learning models have demonstrated remarkable performance in a variety of fields, owing to advancements in computational power and the availability of extensive datasets for training large-scale models. Nonetheless, these models inherently possess a vulnerability wherein even small alterations t...