EditaLive! 面向直播的统一角色视频编辑

摘要

传统视频编辑主要关注场景级内容，而直播则更强调人物主体。然而，将现有视频编辑方法直接应用于以人为中心的直播仍具挑战性，因为它们可能引入面部表情不一致问题，且通常依赖多个离线推理步骤，难以适用于实时交互。我们提出了EditaLive，一个用于实时流式角色视频编辑的新型框架。具体而言，我们从预训练图像动画模型（Wan-Animate）出发，该模型天然地将外观与运动解耦，我们通过参考帧编辑和视频重建，利用收集的CharEdit-50K数据集将其重新改造为基于指令的以人为中心的视频编辑基础模型。此外，我们将模型从离线双向生成适配为因果流式生成，并设计了一种对齐的自rollout蒸馏策略，将模型压缩为两步采样器，其中固定的RoPE和对齐强制减少了训练与推理之间的差异，首帧保留的稀疏注意力过滤了冗余的历史信息，以缓解外观漂移。大量实验表明，EditaLive实现了最先进的编辑性能，能够忠实保留面部表情，并支持低延迟的实时流式推理。

English

Conventional video editing primarily focuses on scene-level content, whereas live streaming places greater emphasis on the human subject. However, directly applying existing video-editing methods to human-centric live streaming remains challenging, as they may introduce facial-expression inconsistencies and typically depend on multiple offline inference steps, making them unsuitable for real-time interaction. We propose EditaLive, a novel framework for real-time streaming character video editing. In detail, we start from a pretrained image animation model (Wan-Animate), which naturally decouples appearance from motion, and repurpose it as the base model for instruction-based human-centric video editing by reference frame editing and video reconstruction via the collected CharEdit-50K dataset. Besides, we adapt the model from offline bidirectional to causal streaming generation, and design an aligned self-rollout distillation strategy that compresses the model into a two-step sampler, where fixed RoPE and align forcing reduce training--inference discrepancies, and first-frame preserved sparse attention filters redundant historical information to mitigate appearance drift. Extensive experiments demonstrate that EditaLive delivers state-of-the-art editing performance with faithful preservation of facial expressions and low-latency real-time streaming inference.