MSLIQA: Enhancing Learning Representations for Image Quality Assessment through Multi-Scale Learning

TL;DR

This paper introduces MSLIQA, a multi-scale learning approach that significantly improves lightweight no-reference image quality assessment models by using novel augmentation strategies and test-time augmentation, achieving performance comparable to state-of-the-art methods.

Contribution

The paper proposes a novel augmentation strategy and test-time augmentation for lightweight NR-IQA models, boosting performance by nearly 28% and enhancing their ability to discriminate image distortions.

Findings

Performance improved by almost 28% with the new augmentation strategy.

Lightweight model results are comparable to state-of-the-art models.

Test-time augmentation further enhances model performance.

Abstract

No-Reference Image Quality Assessment (NR-IQA) remains a challenging task due to the diversity of distortions and the lack of large annotated datasets. Many studies have attempted to tackle these challenges by developing more accurate NR-IQA models, often employing complex and computationally expensive networks, or by bridging the domain gap between various distortions to enhance performance on test datasets. In our work, we improve the performance of a generic lightweight NR-IQA model by introducing a novel augmentation strategy that boosts its performance by almost 28\%. This augmentation strategy enables the network to better discriminate between different distortions in various parts of the image by zooming in and out. Additionally, the inclusion of test-time augmentation further enhances performance, making our lightweight network's results comparable to the current state-of-the-art models, simply through the use of augmentations.