CamPilot：用于摄像机控制电影生成的多智能体电影助手
CamPilot: A Multi-Agent Cinematic Assistant for Camera-Controlled Movie Generation
- Worcester Polytechnic Institute（伍斯特理工学院）
- Adobe Research（Adobe研究院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
CamPilot提出多智能体框架，结合电影规划与摄像机控制，利用GRPO从14K专业电影学习，提升电影生成的连贯性与质量，并建立CamEval基准验证其优越性。
AI中文摘要：
将大语言模型（LLMs）集成到视频生成中，已实现快速的文本到视频生成并提升了视觉质量。然而，这仍达不到专业电影制作的水平，其中电影语言不如人工制作的摄像机工作精细，且多镜头连续性仍具挑战性。为解决这些局限，我们提出了CamPilot，一个多智能体框架，它整合了电影规划与摄像机工作控制，以生成更连贯、逻辑结构更清晰且符合人类审美的电影。CamPilot采用基于GRPO的学习范式，从14K部真实世界专业电影中学习摄像机工作规划，内化运动模式和构图原则，以支持对拍摄技术（如摄像机角度、运动和焦距行为）及跨镜头关系的推理，从而实现可控的摄像机视角生成。多个智能体进一步协作与进化，以提升整体输出质量。为支持这项工作及该领域的进一步研究，我们建立了CamEval，一个用于评估摄像机工作质量和电影参与度的基准。实证结果表明，CamPilot在电影控制和质量上优于最先进的文本到电影生成方法，突显了专业摄像机设计对电影生成的影响。
英文摘要：
The integration of large language models (LLMs) into video generation has enabled rapid text-to-video creation and improved visual quality. However, it still falls short of professional filmmaking, where cinematographic language is less refined than human-crafted camera work and multi-shot continuity remains challenging. To address these limitations, we introduce CamPilot, a multi-agent framework that integrates cinematographic planning and camera-work control to produce more coherent, logically structured, and human-aesthetic movies. CamPilot adopts a GRPO-based learning paradigm to learn camera work planning from 14K real-world professional movies, internalizing motion patterns and composition principles that support reasoning over shooting techniques (e.g., camera angle, motion, and focal behavior) and cross-shot relationships for controllable camera-viewpoint generation. Multiple agents further collaborate and evolve to improve overall output quality. To support this work and further studies in this domain, we establish CamEval, a benchmark for evaluating camera work quality and cinematic engagement. Empirical results show that CamPilot outperforms state-of-the-art text-to-movie generation methods on cinematographic control and quality, highlighting the impact of professional camera design on movie generation.
