arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2512.02793cs.CV
收藏

IC-World:基于上下文的共享世界建模生成

IC-World: In-Context Generation for Shared World Modeling

Nanyang Technological University(南洋理工大学);Goertek Alpha Labs(歌尔声学实验室);Tencent(腾讯)

Fan Wu, Jiacheng Wei, Ruibo Li, Yi Xu, Junyou Li, Deheng Ye, Guosheng Lin

AI总结:

IC-World通过激活大视频模型的上下文生成能力,实现了多视角共享世界建模,通过强化学习和奖励模型提升生成视频的几何和运动一致性。

摘要:

基于视频的世界模型近年来因其能够合成多样化和动态的视觉环境而受到越来越多的关注。在本文中,我们专注于共享世界建模,其中模型从一组输入图像中生成多个视频,每个视频代表同一底层世界在不同相机姿态下的表现。我们提出了IC-World,一种新的生成框架,通过激活大视频模型固有的上下文生成能力,实现所有输入图像的并行生成。我们进一步通过强化学习、群体相对策略优化以及两个提出的新型奖励模型对IC-World进行微调,以强制生成视频集中的场景级几何一致性和物体级运动一致性。广泛的实验表明,IC-World在几何和运动一致性方面均显著优于最先进的方法。到目前为止,这是首次系统探讨基于视频的世界模型的共享世界建模问题。

英文摘要:

Video-based world models have recently garnered increasing attention for their ability to synthesize diverse and dynamic visual environments. In this paper, we focus on shared world modeling, where a model generates multiple videos from a set of input images, each representing the same underlying world in different camera poses. We propose IC-World, a novel generation framework, enabling parallel generation for all input images via activating the inherent in-context generation capability of large video models. We further finetune IC-World via reinforcement learning, Group Relative Policy Optimization, together with two proposed novel reward models to enforce scene-level geometry consistency and object-level motion consistency among the set of generated videos. Extensive experiments demonstrate that IC-World substantially outperforms state-of-the-art methods in both geometry and motion consistency. To the best of our knowledge, this is the first work to systematically explore the shared world modeling problem with video-based world models.

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。

↑