Learning Conditional Knowledge Distillation for Degraded-Reference Image Quality Assessment

TL;DR

This paper introduces a practical degraded-reference IQA method that uses degraded images as references by distilling knowledge from pristine images, enabling effective blind image quality assessment without needing pristine images during testing.

Contribution

It proposes a novel degraded-reference IQA framework that learns a reference space from degraded images by distilling knowledge from pristine images, improving blind image quality assessment.

Findings

Achieves performance close to full-reference IQA methods.

Effective for GAN-based image restoration quality assessment.

Provides a differentiable metric for blind IR models.

Abstract

An important scenario for image quality assessment (IQA) is to evaluate image restoration (IR) algorithms. The state-of-the-art approaches adopt a full-reference paradigm that compares restored images with their corresponding pristine-quality images. However, pristine-quality images are usually unavailable in blind image restoration tasks and real-world scenarios. In this paper, we propose a practical solution named degraded-reference IQA (DR-IQA), which exploits the inputs of IR models, degraded images, as references. Specifically, we extract reference information from degraded images by distilling knowledge from pristine-quality images. The distillation is achieved through learning a reference space, where various degraded images are encouraged to share the same feature statistics with pristine-quality images. And the reference space is optimized to capture deep image priors that are useful for quality assessment. Note that pristine-quality images are only used during training. Our work provides a powerful and differentiable metric for blind IRs, especially for GAN-based methods. Extensive experiments show that our results can even be close to the performance of full-reference settings.