Robust Unpaired Single Image Super-Resolution of Faces

2022/01/22 by Saurabh Goswami, Goswami, Saurabh, Rajagopalan A. N +1
Computer Science · Physics and Astronomy · #Advanced Image Processing Techniques #Advanced Optical Sensing Technologies #Adversarial Robustness in Machine Learning #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences

paper · pdf · doi:10.48550/arxiv.2201.09109

openalex publication_date 2022/01/22 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We propose an adversarial attack for facial class-specific Single Image Super-Resolution (SISR) methods. Existing attacks, such as the Fast Gradient Sign Method (FGSM) or the Projected Gradient Descent (PGD) method, are either fast but ineffective, or effective but prohibitively slow on these networks. By closely inspecting the surface that the MSE loss, used to train such networks, traces under varying degradations, we were able to identify its parameterizable property. We leverage this property to propose an adverasrial attack that is able to locate the optimum degradation (effective) without needing multiple gradient-ascent steps (fast). Our experiments show that the proposed method is able to achieve a better speed vs effectiveness trade-off than the state-of-theart adversarial attacks, such as FGSM and PGD, for the task of unpaired facial as well as class-specific SISR.

Related

• Evaluating Robustness of Deep Image Super-Resolution against Adversarial Attacks
• Superpixel Attack: Enhancing Black-box Adversarial Attack with Image-driven Division Areas
• Adversarial Attacks on Face Detectors using Neural Net based Constrained Optimization
• Powerful Physical Adversarial Examples Against Practical Face Recognition Systems
• Universal Robustness via Median Randomized Smoothing for Real-World Super-Resolution