Fine-Grained HDR Image Quality Assessment From Noticeably Distorted to Very High Fidelity

TL;DR

This paper introduces AIC-HDR2025, a comprehensive HDR image quality dataset with subjective ratings, enabling precise assessment of high-fidelity HDR images and evaluating existing quality metrics.

Contribution

The paper presents the first HDR dataset focused on high-fidelity quality assessment and demonstrates the effectiveness of the AIC-3 methodology for precise subjective ratings.

Findings

AIC-HDR2025 covers subtle perceptual differences in high-fidelity HDR images.

AIC-3 methodology achieves 95% confidence intervals averaging 0.27 at 1 JND.

Objective metrics show varying degrees of correlation with subjective ratings.

Abstract

High dynamic range (HDR) and wide color gamut (WCG) technologies significantly improve color reproduction compared to standard dynamic range (SDR) and standard color gamuts, resulting in more accurate, richer, and more immersive images. However, HDR increases data demands, posing challenges for bandwidth efficiency and compression techniques. Advances in compression and display technologies require more precise image quality assessment, particularly in the high-fidelity range where perceptual differences are subtle. To address this gap, we introduce AIC-HDR2025, the first such HDR dataset, comprising 100 test images generated from five HDR sources, each compressed using four codecs at five compression levels. It covers the high-fidelity range, from visible distortions to compression levels below the visually lossless threshold. A subjective study was conducted using the JPEG AIC-3 test methodology, combining plain and boosted triplet comparisons. In total, 34,560 ratings were collected from 151 participants across four fully controlled labs. The results confirm that AIC-3 enables precise HDR quality estimation, with 95\% confidence intervals averaging a width of 0.27 at 1 JND. In addition, several recently proposed objective metrics were evaluated based on their correlation with subjective ratings. The dataset is publicly available.