# 利用语义地图进行城市尺度跨视角定位

> 原文链接：https://robotics.hellobike.com/papers/2026-07-28-2607-25215
> 频道：papers | 更新日期：2026-07-28 | 标签：arxiv, cs.RO
> 核心摘要：该方法用VLM提取并匹配全景与OpenStreetMap语义地标，结合贝叶斯滤波，在十一环境轨迹上验证城市尺度跨视角定位。

# 利用语义地图进行城市尺度跨视角定位

::: v-pre
> 原文链接：[Leveraging Semantic Maps for City-Scale Cross-View Localization](https://arxiv.org/abs/2607.25215)（arXiv 2607.25215 · 发布日期：2026-07-28）

**Leveraging Semantic Maps for City-Scale Cross-View Localization**

该方法用VLM提取并匹配全景与OpenStreetMap语义地标，结合贝叶斯滤波，在十一环境轨迹上验证城市尺度跨视角定位。

## 核心要点

- 面向未遍历环境定位，利用OpenStreetMap丰富语义，弥补现有方法忽略或压缩语义信息的问题。
- 用VLM从全景提取相关地标并识别对应关系，再蒸馏轻量匹配器为整张地图计算关联。
- 将匹配输出形成观测似然，并随时间用贝叶斯滤波融合，生成位姿估计序列。
- 发布提取语义和评估轨迹数据集，并在项目网页提供代码与数据集。

**验证范围与局限**：验证范围为十一环境的全景与评估轨迹，包含波士顿暴雪和夜间采集全景。

## 英文摘要 (Abstract)

We want robots to localize in previously untraversed environments against commonly available prior data. Rich semantic data available from OpenStreetMap can be useful in this task. However, existing methods either ignore this semantic information, directly matching panoramas and overhead imagery, or dramatically compress the semantic information, working with a small set of fixed classes. To leverage this rich semantic information, two challenges need to be overcome. First, useful semantic information needs to be extracted from the robot's egocentric observations. Second, the observed information must be quickly associated with the large prior semantic map (e.g., up to 628 km^2). We show that VLMs are effective at both extracting relevant landmarks from panoramas, and identifying feasible correspondences between these landmarks and prior overhead landmarks. However, using VLMs to propose all correspondences scales poorly as the number of mapped landmarks increases. Instead, we propose distilling a lightweight matcher from a VLM which computes correspondences for all entities in a map. We use this output to form an observation likelihood which is fused over time with a Bayes filter to create a time series of pose estimates. To support further investigation into generalizable cross-view methods that leverage semantic information, we release a dataset of extracted semantics and evaluation trajectories spanning eleven environments, including panoramas we collected in a snowstorm and at night in Boston. We demonstrate our method, trained on a single city's fair-weather data, generalizes across location, lighting, weather, and other challenges. Code and datasets are available at https://efahnestock.github.io/loci/.

作者：Ethan Fahnestock, Erick Fuentes, Philip R. Osteen, Nicholas Roy
:::

## 原始参考来源 (Sources)

- [arXiv 2607.25215：Leveraging Semantic Maps for City-Scale Cross-View Localization](https://arxiv.org/abs/2607.25215)