DeepWSD: Projecting Degradations in Perceptual Space to Wasserstein Distance in Deep Feature Space

TL;DR

DeepWSD introduces a novel approach to image quality assessment by modeling perceptual degradation through Wasserstein distance in deep feature space, offering improved interpretability and prediction accuracy over traditional methods.

Contribution

It proposes a new statistical distribution-based framework for IQA using Wasserstein distance in deep features, enhancing interpretability and prediction performance.

Findings

DeepWSD outperforms existing IQA models in accuracy.

The method provides better interpretability of distortion effects.

Extensive experiments validate its superiority in quality prediction.

Abstract

Existing deep learning-based full-reference IQA (FR-IQA) models usually predict the image quality in a deterministic way by explicitly comparing the features, gauging how severely distorted an image is by how far the corresponding feature lies from the space of the reference images. Herein, we look at this problem from a different viewpoint and propose to model the quality degradation in perceptual space from a statistical distribution perspective. As such, the quality is measured based upon the Wasserstein distance in the deep feature domain. More specifically, the 1DWasserstein distance at each stage of the pre-trained VGG network is measured, based on which the final quality score is performed. The deep Wasserstein distance (DeepWSD) performed on features from neural networks enjoys better interpretability of the quality contamination caused by various types of distortions and presents an advanced quality prediction capability. Extensive experiments and theoretical analysis show the superiority of the proposed DeepWSD in terms of both quality prediction and optimization.