基于镜头的思考：具有智能体推理的一致性多镜头视频编辑

摘要

尽管生成式AI显著推动了视频编辑的发展，但现有方法主要聚焦于单镜头或短视频片段。使用多指令编辑长视频仍然是一项严峻的挑战。朴素的切分策略（例如固定时长分割）往往导致实体碎片化、严重的编辑幻觉以及时间连续性的破坏。为弥合这一差距，我们提出了多指令多镜头长视频编辑（MMLVE）任务，该任务围绕三个核心目标构建：跨镜头编辑一致性（CSEC）、多指令解耦（MID）和时空结构零破坏（ZDSS）。为应对这三个独特挑战，我们引入了一种智能体编辑框架，利用大语言模型（LLMs）与视觉语言模型（VLMs）的协同作用实现镜头级视频解耦和精确指令解析。此外，为全面评估该任务，我们构建了MMLVE-Bench——一个聚焦于MMLVE的数据集，其特点是包含复杂的真实世界时空动态、高密度异构指令以及稀疏随机的实体分布。我们进一步利用三个聚焦于MMLVE的评估指标来衡量编辑结果的质量。大量实验表明，我们的MMLVE-Agent优于现有的闭源SOTA方法（如Seedance 2.0），成功消除了编辑幻觉，保持了跨镜头编辑一致性，并实现了无缝的时空过渡。

English

While generative AI has significantly advanced video editing, existing methods primarily focus on single-shot or short video clips. Editing long videos with multiple instructions remains a formidable challenge. Naive chunking strategies, e.g., fixed-duration segmentation, often lead to entity fragmentation, severe editing hallucinations, and disrupted temporal continuity. To bridge this gap, we introduce the Multi-Instruction Multi-Shot Long-Video Editing (MMLVE) task, which is structured around three core objectives: Cross-Shot Editing Consistency (CSEC), Multi-Instruction Decoupling (MID), and Zero-Destruction on Spatiotemporal Structure (ZDSS). To tackle these three unique challenges, we introduce an agentic editing framework that leverages the synergy of Large Language Models (LLMs) and Vision-Language Models (VLMs) to achieve shot-level video decoupling and precise instruction parsing. Furthermore, to comprehensively evaluate this task, we construct MMLVE-Bench, which is an MMLVE-focused dataset characterized by complex real-world spatiotemporal dynamics, high-density heterogeneous instructions, and sparse, random entity distributions. Three MMLVE-focused evaluation metrics are further exploited to assess the quality of the editing results. Extensive experiments demonstrate that our MMLVE-Agent outperforms existing closed-source SOTA approaches (e.g., Seedance 2.0), successfully eliminating editing hallucinations, preserving cross-shot editing consistency, and attaining seamless spatiotemporal transitions.