Embodied AI TopConf · ICCV2025

Move to Understand a 3D Scene: Bridging Visual Grounding and Exploration for Efficient and Versatile Embodied Navigation

ICCV2025 / Vision-Language-Navigation Model

VLN导航3D 视觉感知加速/部署
新标签打开 AlphaXiv新标签打开 Project
Move to Understand a 3D Scene: Bridging Visual Grounding and Exploration for Efficient and Versatile Embodied Navigation figure
AlphaXiv 中文论文页面（可滚动查看）