[论文] Any Time, Any Space: A World-State-Centric Framework for Agentic Video...

论文概要

研究领域: CV 作者: Weijie Wang, Haomin Zhang, Ye Li, Akide Liu, Zeyu Zhang, Haoyu Zhao, Yefei He, Donny Y. Chen, Feng Chen, Bohan Zhuang 发布时间: 2026-10-14 arXiv: 2610.20571

中文摘要

长篇代理式视频生成需要将剧本转化为镜头，并在非连续请求之间保持时空一致性。在镜头间传递提示词或最近的帧无法提供重返场景所需的空间参照系，也无法提供重放早期事件所需的时间索引记录。我们提出 Any Time, Any Space（ATAS），将剧本编译为带有显式位置、相机和事件时间引用的镜头请求。生成前，系统从注册静态几何的场景记忆和按场景、事件、时间索引的动态观测事件记忆中检索相关证据。生成后，ATAS 重建并注册该片段，仅在审查和状态检查接受观测时更新记忆，而影片交付不依赖于记忆准入。在四个视频生成器上，ATAS 改进了所有报告的跨镜头一致性指标，并为每个骨干模型取得了最高的综合基准分数。它在目标图像空间重访任务上优于对比代理和 3D 引导方法，在三项事件重捕标准中领先，并在全部六项人工评价标准中获得最高评分。视频结果和代码见项目页面：https://lhmd.top/atas/

原文摘要

Long-form agentic video generation must turn a screenplay into shots that remain spatially and temporally consistent across nonconsecutive requests. Passing prompts or recent frames between shots does not provide a shared spatial frame for revisiting a scene or a time-indexed record for replaying an earlier event. We present Any Time, Any Space (ATAS), which compiles a screenplay into shot requests with explicit location, camera, and event-time references. Before generation, it retrieves relevant evidence from a scene memory of registered static geometry and an event memory of dynamic observations indexed by scene, event, and time. After generation, ATAS reconstructs and registers the clip. It updates memory only when review and state checks accept the observation, while film delivery rema...

#论文 #arXiv #CV #小凯