MultiShotMaster：可控多镜头视频生成框架
MultiShotMaster: A Controllable Multi-Shot Video Generation Framework
通过改进RoPE实现多镜头视频的灵活编排与跨镜头一致性，为视频生成中的镜头控制与参考注入提供了可复用的位置编码方案。
原文依据 We extend a pretrained single-shot model by integrating two novel variants of RoPE.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论当前视频生成技术擅长生成单镜头片段，但难以生成叙事性多镜头视频，这类视频需要灵活的镜头编排、连贯的叙事以及超越文本提示的可控性。为应对这些挑战，我们提出MultiShotMaster，一个高度可控的多镜头视频生成框架。我们通过集成两种新颖的RoPE变体来扩展预训练的单镜头模型。首先，我们引入多镜头叙事RoPE，它在镜头切换处施加显式相位偏移，从而在保持时序叙事顺序的同时实现灵活的镜头编排。其次，我们设计时空位置感知RoPE，以纳入参考标记和视觉定位信号，实现时空定位的参考注入。此外，为克服数据稀缺问题，我们建立了一个自动化数据标注流水线，用于提取多镜头视频、字幕、跨镜头视觉定位信号和参考图像。我们的框架利用内在架构特性来支持多镜头视频生成，具备文本驱动的镜头间一致性、带运动控制的定制主体以及背景驱动的定制场景。镜头数量和时长均可灵活配置。大量实验证明了我们框架的优越性能和出色的可控性。
英文原文摘要 1345 字符
Current video generation techniques excel at single-shot clips but struggle to produce narrative multi-shot videos, which require flexible shot arrangement, coherent narrative, and controllability beyond text prompts. To tackle these challenges, we propose MultiShotMaster, a framework for highly controllable multi-shot video generation. We extend a pretrained single-shot model by integrating two novel variants of RoPE. First, we introduce Multi-Shot Narrative RoPE, which applies explicit phase shift at shot transitions, enabling flexible shot arrangement while preserving the temporal narrative order. Second, we design Spatiotemporal Position-Aware RoPE to incorporate reference tokens and grounding signals, enabling spatiotemporal-grounded reference injection. In addition, to overcome data scarcity, we establish an automated data annotation pipeline to extract multi-shot videos, captions, cross-shot grounding signals and reference images. Our framework leverages the intrinsic architectural properties to support multi-shot video generation, featuring text-driven inter-shot consistency, customized subject with motion control, and background-driven customized scene. Both shot count and duration are flexibly configurable. Extensive experiments demonstrate the superior performance and outstanding controllability of our framework.