InstructVEdit：一种面向指令式视频编辑的整体性方法
InstructVEdit: A Holistic Approach for Instructional Video Editing
AI总结：
针对指令式视频编辑面临的高质量训练数据稀缺、缺乏整体性框架的问题，提出全周期方法InstructVEdit，通过构建数据集工作流、改进模型架构、迭代优化策略，实现了该任务的当前最优性能，适配多样真实场景。
摘要：
由于难以收集大规模、高质量的编辑视频配对数据，根据指令进行视频编辑是一项极具挑战性的任务。这种数据稀缺不仅限制了训练数据的可获得性，还阻碍了对模型架构和训练策略的系统性探索。现有研究虽在视频编辑的特定方面有所改进（例如利用图像编辑技术合成视频数据集，或采用分解式视频编辑训练），但能应对上述挑战的整体性框架仍未得到充分研究。本研究提出InstructVEdit，这是一种全周期指令式视频编辑方法，包含三方面内容：（1）建立可靠的数据集构建工作流以初始化训练；（2）引入两项模型架构改进，在保持时间一致性的同时提升编辑质量；（3）提出一种利用真实世界数据的迭代优化策略，以增强泛化能力并最小化训练-测试差异。大量实验表明，InstructVEdit在基于指令的视频编辑任务上取得了当前最优性能，展现出对多样真实场景的强适应性。项目主页：https://o937-blip.github.io/InstructVEdit。
英文摘要：
Video editing according to instructions is a highly challenging task due to the difficulty in collecting large-scale, high-quality edited video pair data. This scarcity not only limits the availability of training data but also hinders the systematic exploration of model architectures and training strategies. While prior work has improved specific aspects of video editing (e.g., synthesizing a video dataset using image editing techniques or decomposed video editing training), a holistic framework addressing the above challenges remains underexplored. In this study, we introduce InstructVEdit, a full-cycle instructional video editing approach that: (1) establishes a reliable dataset curation workflow to initialize training, (2) incorporates two model architectural improvements to enhance edit quality while preserving temporal consistency, and (3) proposes an iterative refinement strategy leveraging real-world data to enhance generalization and minimize train-test discrepancies. Extensive experiments show that InstructVEdit achieves state-of-the-art performance in instruction-based video editing, demonstrating robust adaptability to diverse real-world scenarios. Project page: https://o937-blip.github.io/InstructVEdit.
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。
