ICCV 2025 / 神经仿真 / 场景重建 / 占用表示 / 自回归仿真

InfiniCube: Unbounded and Controllable Dynamic 3D Driving Scene Generation with World-Guided Video Models

作者：Yifan Lu, Xuanchi Ren, Jiawei Yang, Tianchang Shen, Zhangjie Wu, Jun Gao, Yue Wang, Siheng Chen, Mike Chen, Sanja Fidler, Jiahui Huang · 单位：NVIDIA, Shanghai Jiao Tong University, University of Toronto, Vector Institute, University of Southern California · 会议/期刊：ICCV 2025 · 方向：神经仿真 / 场景重建 / 占用表示 / 自回归仿真 · 发布日期：2025

世界模型3D/4D建模视频生成占用预测自动驾驶
新标签打开 AlphaXiv新标签打开 Project新标签打开 Code
InfiniCube: Unbounded and Controllable Dynamic 3D Driving Scene Generation with World-Guided Video Models figure
AlphaXiv 中文概览（可滚动查看）