[论文] From Pixels to States: Rethinking Interactive World Models as Game Eng…

## 论文概要 **研究领域**: CV **作者**: Zhen Li, Zian Meng, Shuwei...

论文概要

研究领域: CV 作者: Zhen Li, Zian Meng, Shuwei Shi, Mingliang Zhai, Jiaming Tan, Chuanhao Li, Kaipeng Zhang 发布时间: 2026-07-15 arXiv: 2607.14076

中文摘要

构建对玩家动作做出连贯响应的交互式世界一直是计算机图形学、游戏和人工智能的共同目标。近期视频生成模型通过预测以用户动作为条件的未来观测,为此目标提供数据驱动路径,并越来越被视为潜在的下一代游戏引擎。然而,实现真正交互式游戏世界需要遵循演变游戏条件规则的交互结果、长时段持续的结果,以及实时操作的生成循环。传统游戏引擎通过循环的动作-状态-观测循环实现这些属性,其中玩家动作根据预定义规则更新显式游戏状态,观测从结果状态渲染。以此循环为组织视角,本文从四个维度审视交互式游戏世界建模:玩家动作控制、游戏状态动态、状态-观测持久性和实时交互生成。对每个维度,我们从交互式游戏世界所需能力出发,将现有方法分组为代表性家族,并讨论每个家族的优势和权衡。作为补充分析,我们提出《黑神话:悟空》的可扩展数据引擎,收集超过90小时的游戏玩法,包含帧对齐玩家动作、真实游戏状态和视觉观测,以及结构和语义注释,作为状态感知游戏世界建模的资源。我们希望本文能提供该领域现状的清晰图景,并促进交互式游戏世界的进展。

原文摘要

Building interactive worlds that respond coherently to player actions has long been a shared goal of computer graphics, games, and artificial intelligence. Recent video generative models provide a data-driven route toward this goal by predicting future observations conditioned on user actions, and are increasingly regarded as potential next-generation game engines. Realizing a genuinely interactive game world, however, requires interaction outcomes that follow rules over evolving game conditions, consequences that persist over long horizons, and a generation loop that operates in real time. Conventional game engines realize these properties through a recurrent action-state-observation loop, in which player actions update an explicit game state according to predefined rules and observations…

自动采集于 2026-07-17

#论文 #arXiv #CV #小凯

发表回复

人生梦想 - 关注前沿的计算机技术 acejoy.com 🐾 步子哥の博客 🐾 背多分论坛 🐾 借一步网 🐾 智柴网 沪ICP备2024052574号-1