Image Quality Assessment: Enhancing Perceptual Exploration and Interpretation with Collaborative Feature Refinement and Hausdorff distance

TL;DR

This work introduces a training-free FR-IQA method aligned with human visual perception by combining a wavelet-based collaborative feature refinement module with a Hausdorff distance-based distribution similarity measurement. The approach captures low-frequency distortions (color/luminance) via low-frequency wavelet subbands and high-frequency distortions (edges/textures) via high-frequency subbands, while robustly comparing feature distributions through the Hausdorff distance and supporting color-histogram similarity. The final quality score $Q_p$ fuses these components without reliance on subjective scores, and extensive experiments on LIVE, CSIQ, TID2008/2013, and KADID-10k demonstrate state-of-the-art performance and strong HVS correlation. The training-free nature enhances generalizability, reduces data requirements, and offers a practical tool for perceptual image quality assessment in real-world computer vision tasks.

Abstract

Current full-reference image quality assessment (FR-IQA) methods often fuse features from reference and distorted images, overlooking that color and luminance distortions occur mainly at low frequencies, whereas edge and texture distortions occur at high frequencies. This work introduces a pioneering training-free FR-IQA method that accurately predicts image quality in alignment with the human visual system (HVS) by leveraging a novel perceptual degradation modelling approach to address this limitation. First, a collaborative feature refinement module employs a carefully designed wavelet transform to extract perceptually relevant features, capturing multiscale perceptual information and mimicking how the HVS analyses visual information at various scales and orientations in the spatial and frequency domains. Second, a Hausdorff distance-based distribution similarity measurement module robustly assesses the discrepancy between the feature distributions of the reference and distorted images, effectively handling outliers and variations while mimicking the ability of HVS to perceive and tolerate certain levels of distortion. The proposed method accurately captures perceptual quality differences without requiring training data or subjective quality scores. Extensive experiments on multiple benchmark datasets demonstrate superior performance compared with existing state-of-the-art approaches, highlighting its ability to correlate strongly with the HVS.\footnote{The code is available at \url{https://anonymous.4open.science/r/CVPR2025-F339}.}

Figures (2)

• Figure 1: Framework of the proposed FR-IQA algorithm.
• Figure 2: Visualization of the different methods across databases.