[论文] GeoState: Learning View-Decoupled Geometry States for 3D World Modelin...

论文概要

研究领域: CV
作者: Hongchang Liu, Jian Liu, Yujuan Ding, Liang Peng, Xun Jiang, Fei Ma, Xing Xu, Yi Bin
发布时间: 2026-10-15
arXiv: 2610.22343

中文摘要

最近的 3D 世界模型越来越多地使用几何基础模型（GFM）的特征作为预测性世界状态，受益于其丰富的几何先验。然而，这些特征的视角依赖性将场景动态与相机引起的变化纠缠在一起。我们提出 GeoState，这是一个视角解耦几何表示，将场景状态演化与相机依赖的观测分离。场景 tokenizer 将多视角 GFM 特征聚合为视角解耦的场景状态，几何解码器使用相机射线查询该状态。自回归世界模型预测未来场景状态，而相机运动单独建模。预测的状态因此可以在预测或外部指定的相机下解码，同时在整个 rollout 中保持单一上下文推导的尺度。实验表明，GeoState 在 Waymo 上优于先前的 3D 世界模型，支持几何一致的跨视角查询，并零样本泛化到 KITTI 和 nuScenes。项目页面：https://hongchang-liu.github.io/GeoState-project-page/

原文摘要

Recent 3D world models increasingly use features from geometry foundation models (GFMs) as predictive world states, benefiting from their rich geometric priors. However, the view dependence of these features entangles scene dynamics with camera-induced variation, making the resulting state evolution harder to predict. We propose GeoState, a view-decoupled geometry representation for predictive 3D world modeling that separates scene-state evolution from camera-dependent observation. Specifically, a scene tokenizer aggregates multi-view GFM features into a view-decoupled scene state, which a geometry decoder queries using camera rays. An autoregressive world model then predicts future scene states, while camera motion is modeled separately. The predicted states can therefore be decoded under...

自动采集于 2026-10-17

#论文 #arXiv #CV #小凯