让机器人精准定位到目标图像，实现厘米级导航。
AnyImageNav: Any-View Geometry for Precise Last-Meter Image-Goal Navigation
- 用语义引导探索，仅在相关时调用3D多视角模型进行几何匹配。
- 在Gibson和HM3D数据集上成功率达93.1%和82.6%，位置误差低于0.3m。
- 适合需要精确对齐的任务，如抓取、装配等高精度场景。
图像目标导航（ImageNav）通常以1米内到达为目标，适用于寻找物体，但无法满足抓取等下游任务对精确定位的需求。本文提出AnyImageNav，一种无需训练的系统，将导航推向更严格的设定。核心思想是将目标图像视为几何查询：任何物体、走廊或房间角落的照片均可通过密集像素对应与代理观测对齐，从而恢复精确的6-自由度相机位姿。方法采用语义到几何的级联流程：语义相关性信号引导探索并作为接近门控，仅在当前视图与目标高度相关时激活3D多视角基础模型；模型通过循环自认证实现注册，获得准确位姿。在Gibson数据集上取得93.1%的成功率，位置误差0.27米，朝向误差3.41度；在HM3D上成功率达82.6%，位置误差0.21米，朝向误差1.23度，相较适配基线提升5-10倍。项目主页：https://yijie21.github.io/ain/
原文摘要 · Abstract (English)
Image Goal Navigation (ImageNav) is evaluated by a coarse success criterion, the agent must stop within 1m of the target, which is sufficient for finding objects but falls short for downstream tasks such as grasping that require precise positioning. We introduce AnyImageNav, a training-free system that pushes ImageNav toward this more demanding setting. Our key insight is that the goal image can be treated as a geometric query: any photo of an object, a hallway, or a room corner can be registered to the agent's observations via dense pixel-level correspondences, enabling recovery of the exact 6-DoF camera pose. Our method realizes this through a semantic-to-geometric cascade: a semantic relevance signal guides exploration and acts as a proximity gate, invoking a 3D multi-view foundation model only when the current view is highly relevant to the goal image; the model then self-certifies its registration in a loop for an accurate recovered pose. Our method sets state-of-the-art navigation success rates on Gibson (93.1%) and HM3D (82.6%), and achieves pose recovery that prior methods do not provide: a position error of 0.27m and heading error of 3.41 degrees on Gibson, and 0.21m / 1.23 degrees on HM3D, a 5-10x improvement over adapted baselines.Our project page: https://yijie21.github.io/ain/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。