让图像质量评估模型像人一样感知并推理，提升可解释性。
Guiding Perception-Reasoning Closer to Human in Blind Image Quality Assessment
- 用强化学习和人类标注作为奖励信号，引导模型模仿人类感知-推理过程。
- 在超过1000个样本上，模型生成的推理链与人类匹配度达ROUGE-1 0.512，优于基线的0.443。
- 不仅预测准，还能自动生成类似人类的解释，适合需要可解释性的评估场景。
人类评估图像质量依赖感知与推理的连续过程，结合感官线索与隐含判断形成自洽结论。本文研究如何使盲图像质量评估（BIQA）模型具备类人且自洽的推理能力。首先，收集包含人类感知-推理多阶段的评估数据；随后，采用强化学习，以人类标注为奖励信号，引导模型向类人感知与推理靠拢。为实现自洽推理，设计一种奖励机制，促使模型仅基于自生成描述推断图像质量。实验表明，该方法在通用指标上达到与当前最优BIQA系统相当的性能，包括皮尔逊与斯皮尔曼相关系数。此外，通过ROUGE-1衡量模型生成推理链与人类链的相似性，在超过1000个标注样本上，模型达到0.512的得分（基线为0.443），显著覆盖人类解释，标志着在可解释性方面向类人推理迈出关键一步。
原文摘要 · Abstract (English)
Humans assess image quality through a perception-reasoning cascade, integrating sensory cues with implicit reasoning to form self-consistent judgments. In this work, we investigate how a model can acquire both human-like and self-consistent reasoning capability for blind image quality assessment (BIQA). We first collect human evaluation data that capture several aspects of human perception-reasoning pipeline. Then, we adopt reinforcement learning, using human annotations as reward signals to guide the model toward human-like perception and reasoning. To enable the model to internalize self-consistent reasoning capability, we design a reward that drives the model to infer the image quality purely from self-generated descriptions. Empirically, our approach achieves score prediction performance comparable to state-of-the-art BIQA systems under general metrics, including Pearson and Spearman correlation coefficients. In addition to the rating score, we assess human-model alignment using ROUGE-1 to measure the similarity between model-generated and human perception-reasoning chains. On over 1,000 human-annotated samples, our model reaches a ROUGE-1 score of 0.512 (cf. 0.443 for baseline), indicating substantial coverage of human explanations and marking a step toward human-like interpretable reasoning in BIQA.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。