TDVE-Assessor：利用大语言模型（LMMs）基准测试与评估文本驱动视频编辑质量
TDVE-Assessor: Benchmarking and Evaluating the Quality of Text-Driven Video Editing with LMMs
AI总结：
针对文本驱动视频编辑缺乏专用质量评估模型的问题，构建大规模基准TDVE-DB并评估现有方法，进而提出融合时空特征与LLM的TDVE-Assessor，在三个维度上均达到最优性能。
摘要：
文本驱动的视频编辑技术正在快速发展，然而由于缺乏能够辨别编辑质量细微差别的专用视频质量评估（VQA）模型，其严格的评估仍然具有挑战性。为解决这一关键空白，我们引入了TDVE-DB，一个用于文本驱动视频编辑的大规模基准数据集。TDVE-DB包含由8个编辑类别中的12个不同模型生成的3,857个编辑视频，并沿三个关键维度（即编辑视频质量、编辑对齐和结构一致性）标注了173,565个人类主观评分。基于TDVE-DB，我们首先对12个最先进的编辑模型进行了全面评估，揭示了当前视频技术的优缺点，然后在文本驱动视频编辑评估的背景下对现有VQA方法进行了基准测试。基于这些见解，我们提出了TDVE-Assessor，一种专为文本驱动视频编辑评估设计的新型VQA模型。TDVE-Assessor将空间和时间视频特征整合到大语言模型（LLM）中，以提供丰富的上下文理解，从而提供全面的质量评估。大量实验表明，TDVE-Assessor在TDVE-DB上所有三个评估维度均显著优于现有VQA模型，树立了新的最先进水平。TDVE-DB和TDVE-Assessor将在论文发表后发布。
英文摘要：
Text-driven video editing is rapidly advancing, yet its rigorous evaluation remains challenging due to the absence of dedicated video quality assessment (VQA) models capable of discerning the nuances of editing quality. To address this critical gap, we introduce TDVE-DB, a large-scale benchmark dataset for text-driven video editing. TDVE-DB consists of 3,857 edited videos generated from 12 diverse models across 8 editing categories, and is annotated with 173,565 human subjective ratings along three crucial dimensions, i.e., edited video quality, editing alignment, and structural consistency. Based on TDVE-DB, we first conduct a comprehensive evaluation for the 12 state-of-the-art editing models revealing the strengths and weaknesses of current video techniques, and then benchmark existing VQA methods in the context of text-driven video editing evaluation. Building on these insights, we propose TDVE-Assessor, a novel VQA model specifically designed for text-driven video editing assessment. TDVE-Assessor integrates both spatial and temporal video features into a large language model (LLM) for rich contextual understanding to provide comprehensive quality assessment. Extensive experiments demonstrate that TDVE-Assessor substantially outperforms existing VQA models on TDVE-DB across all three evaluation dimensions, setting a new state-of-the-art. Both TDVE-DB and TDVE-Assessor will be released upon the publication.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
