arXiv cs.AI· Hongchang Liu, Jian Liu, Yujuan Ding, Liang Peng, Xun Jiang, Fei Ma, Xing Xu, Yi Bin·· 2 天前AI 评分25
GeoState：面向 3D 世界建模的视角解耦几何状态学习
GeoState: Learning View-Decoupled Geometry States for 3D World Modeling
AI 导读
GeoState 是一种视角解耦的几何表示，用于预测式 3D 世界建模，将场景状态演化与相机相关观测分离。它通过场景 tokenizer 将多视角 GFM 特征聚合为视角解耦的场景状态，由几何解码器用相机射线查询，自回归世界模型预测未来场景状态，相机运动单独建模。
正文 · 原文
Abstract: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 either predicted or externally specified cameras, while retaining a single context-derived scale throughout the rollout without per-frame scale alignment. Experiments show that GeoState outperforms prior 3D world models on Waymo, supports geometrically consistent cross-view querying, and generalizes zero-shot to KITTI and nuScenes. Project page: this https URL
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Robotics (cs.RO) |
| Cite as: | arXiv:2610.22343 [cs.CV] |
| (or arXiv:2610.22343v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.22343 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Hongchang Liu [view email]
[v1]
Thu, 15 Oct 2026 17:56:18 UTC (2,685 KB)
来源：arXiv cs.AI · arxiv.org