提出新方法提升多维度视频质量评分准确率
Analytic Score Optimization for Multi Dimension Video Quality Assessment
- 将质量评估建模为带约束的决策过程，获得闭式解
- 在五个维度上优于主流模型，降低评分误差
- 适合需要可解释性视频评估的开发者与研究者
视频质量评估正从单一得分转向更丰富的多维评价。本文构建了大规模多维度视频质量数据集UltraVQA，涵盖用户生成内容（UGC），标注五个关键维度：运动质量、运动幅度、美学质量、内容质量和清晰度质量。每段视频由超过3名人类评分员打分，并配有基于集体判断由GPT生成的解释性理由。为更好利用这些丰富标注，我们提出分析型评分优化（ASO），一种理论基础扎实的后训练目标，将质量评估重新定义为带正则化的决策过程，得到闭式解，自然捕捉人类评分的序数特性，确保与人类排序偏好一致。实验表明，该方法在多数基线（包括闭源API和开源模型）中表现更优，同时降低质量预测的平均绝对误差（MAE）。本工作凸显了多维、可解释标注及基于强化的学习对视频质量评估的重要性。
原文摘要 · Abstract (English)
Video Quality Assessment (VQA) is evolving beyond single-number mean opinion score toward richer, multi-faceted evaluations of video content. In this paper, we present a large-scale multi-dimensional VQA dataset UltraVQA that encompasses diverse User-Generated Content~(UGC) annotated across five key quality dimensions: Motion Quality, Motion Amplitude, Aesthetic Quality, Content Quality, and Clarity Quality. Each video in our dataset is scored by over 3 human raters on these dimensions, with fine-grained sub-attribute labels, and accompanied by an explanatory rationale generated by GPT based on the collective human judgments. To better leverage these rich annotations and improve discrete quality score assessment, we introduce Analytic Score Optimization (ASO), a theoretically grounded post-training objective derived for multi-dimensional VQA. By reframing quality assessment as a regularized decision-making process, we obtain a closed-form solution that naturally captures the ordinal nature of human ratings, ensuring alignment with human ranking preferences. In experiments, our method outperforms most baselines including closed-source APIs and open-source models, while also reducing mean absolute error (MAE) in quality prediction. Our work highlights the importance of multi-dimensional, interpretable annotations and reinforcement-based alignment in advancing video quality assessment.
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。