无需训练即可实现多种视频编辑，保持细节与身份一致性。
One Editor, Many Edits: A Unified Training-Free Framework for Diverse Video Editing
- 用稀疏因果记忆保持局部连贯性，通过特征注入维持长程身份一致。
- 在FiVE数据集上达到78.16分，远超最强基线58.95分。
- 支持指令和参考引导编辑，适合视频创作与个性化修改场景。
视频编辑涵盖多种范式，但如何在单一统一框架中实现高质量的指令引导与主体引导编辑仍具挑战。本文提出EditVid，一种无需训练的框架，结合稀疏因果记忆以保持局部连贯性、基于对应关系的后注意力令牌注入以维持长距离身份一致性，以及软潜在融合以控制编辑局部性。该框架可统一支持指令引导与参考引导编辑，包括风格迁移、属性修改、物体插入、局部编辑及主体替换。在FiVE数据集上，EditVid取得78.16的FiVE-Acc，显著优于最强基线（58.95），并在IVEBench上表现具有竞争力。用户研究显示，其整体偏好度达51.8%，优于7种对比方法。
原文摘要 · Abstract (English)
Video editing spans diverse editing paradigms, yet achieving high-quality instruction-guided and subject-guided editing within a single unified framework remains challenging. We introduce EditVid, a training-free framework combining sparse causal memory for local coherence, correspondence-based post-attention token injection for long-range identity preservation, and soft latent blending for edit locality. The same framework supports instruction-guided and reference-guided edits, including style transfer, attribute modification, object insertion, part-level editing, and subject replacement. On FiVE, EditVid achieves 78.16 FiVE-Acc, compared with 58.95 for the strongest evaluated training-free baseline, while obtaining competitive results on IVEBench. A user study further shows a 51.8\% overall preference for EditVid over 7 competing methods.
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。