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

arXiv: https://arxiv.org/abs/2607.25215 · HTML: https://mlsanity.com/paper/2607.25215/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 33 references that are in the map.
- First posted on arXiv 28 Jul 2026.

## Builds on (most cited first, 19 of 19)

- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Gemma 3 Technical Report](https://mlsanity.com/paper/2503.19786/) (2025)
- [SuperGlue: Learning Feature Matching with Graph Neural Networks](https://mlsanity.com/paper/1911.11763/) (2019)
- [VIGOR: Cross-View Image Geo-localization beyond One-to-one Retrieval](https://mlsanity.com/paper/2011.12172/) (2020)
- [Wide-Area Image Geolocalization with Aerial Reference Imagery](https://mlsanity.com/paper/1510.03743/) (2015)
- [TransGeo: Transformer Is All You Need for Cross-view Image Geo-localization](https://mlsanity.com/paper/2204.00097/) (2022)
- [OrienterNet: Visual Localization in 2D Public Maps with Neural Matching](https://mlsanity.com/paper/2304.02009/) (2023)
- [Cross-View Image Matching for Geo-localization in Urban Environments](https://mlsanity.com/paper/1703.07815/) (2017)
- [Where am I? Cross-View Geo-localization with Natural Language Descriptions](https://mlsanity.com/paper/2412.17007/) (2024)
- [Entity Matching using Large Language Models](https://mlsanity.com/paper/2310.11244/) (2023)
- [Image-Based Geo-Localization Using Satellite Imagery](https://mlsanity.com/paper/1903.00159/) (2019)
- [CV-Cities: Advancing Cross-View Geo-Localization in Global Cities](https://mlsanity.com/paper/2411.12431/) (2024)
- [You Are Here: Geolocation by Embedding Maps and Images](https://mlsanity.com/paper/1911.08797/) (2019)
- [Cross-view geo-localization: a survey](https://mlsanity.com/paper/2406.09722/) (2024)
- [City-wide Street-to-Satellite Image Geolocalization of a Mobile Ground Agent](https://mlsanity.com/paper/2203.05612/) (2022)
- [Scaling Image Geo-Localization to Continent Level](https://mlsanity.com/paper/2510.26795/) (2025)
- [TextSLAM: Visual SLAM with Semantic Planar Text Features](https://mlsanity.com/paper/2305.10029/) (2023)
- [SegLocNet: Multimodal Localization Network for Autonomous Driving via Bird's-Eye-View Segmentation](https://mlsanity.com/paper/2502.20077/) (2025)
- [Text Detection & Recognition in the Wild for Robot Localization](https://mlsanity.com/paper/2205.08565/) (2022)

## Abstract (arXiv)

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/.