FS-IQA: Certified Feature Smoothing for Robust Image Quality Assessment

TL;DR

This paper introduces a feature space randomized smoothing technique for robust image quality assessment that maintains image fidelity and offers formal robustness guarantees, significantly improving efficiency and accuracy.

Contribution

It presents a novel certified defense method for IQA models using feature space noise, reducing inference time and enhancing robustness without modifying model architecture.

Findings

Achieves up to 30.9% improvement in correlation with subjective scores.

Reduces inference time by 99.5% without certification.

Supports both full-reference and no-reference IQA models.

Abstract

We propose a novel certified defense method for Image Quality Assessment (IQA) models based on randomized smoothing with noise applied in the feature space rather than the input space. Unlike prior approaches that inject Gaussian noise directly into input images, often degrading visual quality, our method preserves image fidelity while providing robustness guarantees. To formally connect noise levels in the feature space with corresponding input-space perturbations, we analyze the maximum singular value of the backbone network's Jacobian. Our approach supports both full-reference (FR) and no-reference (NR) IQA models without requiring any architectural modifications, suitable for various scenarios. It is also computationally efficient, requiring a single backbone forward pass per image. Compared to previous methods, it reduces inference time by 99.5% without certification and by 20.6% when certification is applied. We validate our method with extensive experiments on two benchmark datasets, involving six widely-used FR and NR IQA models and comparisons against five state-of-the-art certified defenses. Our results demonstrate consistent improvements in correlation with subjective quality scores by up to 30.9%.