Deep Quality: A Deep No-reference Quality Assessment System

TL;DR

Deep Quality introduces a novel deep learning-based no-reference image quality assessment system that effectively predicts perceived image quality without reference images, achieving performance comparable to full-reference methods.

Contribution

It proposes a multi-scale deep convolutional neural network for no-reference IQA, addressing a longstanding challenge with a novel deep learning approach.

Findings

Achieved 89% patch-level accuracy on CSIQ dataset.

Achieved 98% image-level accuracy on CSIQ dataset.

Performs comparably to full-reference IQA methods.

Abstract

Image quality assessment (IQA) continues to garner great interest in the research community, particularly given the tremendous rise in consumer video capture and streaming. Despite significant research effort in IQA in the past few decades, the area of no-reference image quality assessment remains a great challenge and is largely unsolved. In this paper, we propose a novel no-reference image quality assessment system called Deep Quality, which leverages the power of deep learning to model the complex relationship between visual content and the perceived quality. Deep Quality consists of a novel multi-scale deep convolutional neural network, trained to learn to assess image quality based on training samples consisting of different distortions and degradations such as blur, Gaussian noise, and compression artifacts. Preliminary results using the CSIQ benchmark image quality dataset showed that Deep Quality was able to achieve strong quality prediction performance (89% patch-level and 98% image-level prediction accuracy), being able to achieve similar performance as full-reference IQA methods.