发表机构
Adobe Research; Johns Hopkins University（Adobe研究院; 约翰·霍普金斯大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
Wonder是用于实时相机可控世界探索的通用视频世界模型，通过系统级协同设计，包括新型相机条件设定、高效内存机制等，能合成多样视频，支持视频条件生成，保持长时间的连贯视觉效果。
AI 中文摘要
我们展示了Wonder，一种用于实时、相机可控世界探索的通用视频世界模型。给定图像或条件视频，Wonder构建一个可玩的世界，用户能通过移动相机交互式导航，实时且长期地发现未见区域并回访先前观察区域。实现此能力需要控制方法、内存机制和训练策略的系统级协同设计。我们引入一种带密集坐标场的新型相机条件设定，其渲染提供空间对齐的运动和方向线索。还提出基于高效稀疏注意力的内存机制，并开发多种技术纠正自强制式蒸馏管道，使Wonder能以16帧每秒合成多样的分钟级视频，同时跨长时间保持连贯的几何、外观和动态。此外，Wonder自然支持视频条件生成，可实时重新拍摄现有动态场景。
英文摘要
We present Wonder, a general-purpose video world model for real-time, camera-controllable world exploration. Given an image or a conditional video, Wonder constructs a playable world where users can navigate interactively by moving the camera, discovering unseen regions, and revisiting previously observed areas in real time and over a long-term horizon. Achieving this capability requires a system-level co-design of control method, memory mechanism, and training strategy. We introduce a novel camera conditioning with a dense coordinate field whose renderings provide spatially aligned motion and orientation cues, allowing the model to interpret camera motion directly as visual evidence. To support fast and precise memory retrieval over a growing generation context, we propose an efficient sparse attention-based memory mechanism, enabling the model to selectively attend to a small set of relevant context tokens at inference time, regardless of actual context length. We further develop several techniques to rectify the self-forcing-style distillation pipeline, improving the student model's ability to respect control signals, as well as maintaining diverse generation modes and long-term memory from the teacher. Together, these components enable Wonder to synthesize diverse, minute-scale videos at 16 FPS while preserving coherent geometry, appearance, and dynamics across long rollouts. Beyond image-to-video generation, Wonder naturally supports video-conditioned generation, allowing existing dynamic scenes to be re-shot in real time.
CommentsProject Page: https://wonder-world-model.github.io/
