弥合图像超分辨率评测中的感知鸿沟
Bridging the Perception Gap in Image Super-Resolution Evaluation
对做超分辨率的人，这篇给出了一个可直接套用的评测改造：把任意 FR/NR IQA 指标换成成对相对质量比较（RQI），既能更贴合人眼偏好地排名 SR 模型，也能当作训练损失引导生成更真实的细节。
原文依据 we propose a simple yet effective Relative Quality Index (RQI) framework, which assesses the relative quality discrepancy between image pairs
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论随着超分辨率（SR）技术的进步，我们注意到近期 SR 研究中对评测指标的不信任日益增长。某些评测标准与人类感知偏好之间经常出现不一致。尽管当前 SR 研究采用不同的指标来评估 SR 性能，但这些指标究竟有多稳健、多可靠仍缺乏充分探讨。为弥合这一鸿沟，我们对广泛使用的图像质量指标进行了全面分析，考察其在评估最优 SR 模型时与人类感知的一致性。我们表明，部分指标与人类偏好仅呈有限相关，甚至呈负相关。我们进一步指出 SR 评测中若干内在挑战，它们削弱了全参考（FR）与无参考（NR）图像质量评估（IQA）框架的有效性。为解决这些问题，我们提出一种简单而有效的相对质量指数（RQI）框架，它评估图像对之间的相对质量差异。该框架能够便捷地集成到现有 IQA 指标中，并在 SR 评测中带来显著提升。此外，它还可作为 SR 模型有价值的训练引导，使生成的图像在保持结构保真度的同时具有更真实的细节。
英文原文摘要 1309 字符
As super-resolution (SR) techniques advance, we observe a growing distrust of evaluation metrics in recent SR research. An inconsistency often emerges between certain evaluation criteria and human perceptual preference. Although current SR research employs varying metrics to evaluate SR performance, it remains underexplored how robust and reliable these metrics actually are. To bridge this gap, we conduct a comprehensive analysis of widely used image quality metrics, examining their consistency with human perception when evaluating state-of-the-art SR models. We show that some metrics exhibit only limited--or even negative--correlation with human preferences. We further identify several intrinsic challenges in SR evaluation that compromise the effectiveness of both full-reference (FR) and no-reference (NR) image quality assessment (IQA) frameworks. To address these issues, we propose a simple yet effective Relative Quality Index (RQI) framework, which assesses the relative quality discrepancy between image pairs. Our framework enables easy integration and notable improvements for existing IQA metrics in SR evaluation. Moreover, it can be utilized as a valuable training guide for SR models, enabling the generation of images with more realistic details while maintaining structural fidelity.