MultiShotMaster: 一种可控的多镜头视频生成框架
MultiShotMaster: A Controllable Multi-Shot Video Generation Framework
AI总结：
MultiShotMaster通过引入两种新型RoPE变体和自动化数据标注管道，实现多镜头视频的灵活生成与可控性。
摘要：
当前视频生成技术在单镜头片段方面表现出色，但难以生成具有叙事性的多镜头视频，这些视频需要灵活的镜头安排、连贯的叙事和超越文本提示的可控性。为解决这些问题，我们提出了MultiShotMaster，一种高度可控的多镜头视频生成框架。我们通过整合两种新的RoPE变体扩展了预训练的单镜头模型。首先，我们引入了Multi-Shot Narrative RoPE，在镜头过渡时应用显式的相位移，使镜头安排灵活的同时保持时间叙事顺序。其次，我们设计了时空位置感知RoPE，以整合参考标记和基础信号，使时空基础的参考注入成为可能。此外，为克服数据稀缺问题，我们建立了自动化数据标注管道，以提取多镜头视频、说明、跨镜头基础信号和参考图像。我们的框架利用内在的架构属性支持多镜头视频生成，具有文本驱动的跨镜头一致性、定制的主体与运动控制以及背景驱动的定制场景。镜头数量和持续时间都可以灵活配置。大量实验表明，我们的框架具有优越的性能和出色的可控性。
英文摘要：
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.
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。
