DDR: Exploiting Deep Degradation Response as Flexible Image Descriptor

TL;DR

DDR introduces a novel method to quantify and utilize deep image features under various degradation conditions, serving as a versatile descriptor for quality assessment and image restoration tasks.

Contribution

The paper presents Deep Degradation Response (DDR), a new approach that leverages degradation-induced changes in deep features for flexible image analysis and enhancement.

Findings

DDR correlates strongly with image attributes like sharpness and colorfulness.

DDR outperforms existing blind image quality assessment methods.

DDR improves performance in image deblurring and super-resolution tasks.

Abstract

Image deep features extracted by pre-trained networks are known to contain rich and informative representations. In this paper, we present Deep Degradation Response (DDR), a method to quantify changes in image deep features under varying degradation conditions. Specifically, our approach facilitates flexible and adaptive degradation, enabling the controlled synthesis of image degradation through text-driven prompts. Extensive evaluations demonstrate the versatility of DDR as an image descriptor, with strong correlations observed with key image attributes such as complexity, colorfulness, sharpness, and overall quality. Moreover, we demonstrate the efficacy of DDR across a spectrum of applications. It excels as a blind image quality assessment metric, outperforming existing methodologies across multiple datasets. Additionally, DDR serves as an effective unsupervised learning objective in image restoration tasks, yielding notable advancements in image deblurring and single-image super-resolution. Our code is available at: https://github.com/eezkni/DDR