SIO-Mapper: A Framework for Lane-Level HD Map Construction Using Satellite Images and OpenStreetMap with No On-Site Visits

2025-04-14

AI Summary

This work addresses the limitations of conventional high-definition (HD) map construction—namely, its reliance on costly, labor-intensive field data collection, resulting in limited geographic coverage and scalability. We propose a fully automated, city-scale lane-level mapping method leveraging only satellite imagery and OpenStreetMap (OSM) data. Methodologically, we introduce SIO-Net, a multi-source feature fusion network integrating dual encoders (Transformer and CNN), coupled with a cluster-graph collaborative lane integration strategy to ensure semantic consistency and seamless stitching of large-scale lane structures. Evaluated on Naver Labs Open and NuScenes datasets (covering South Korea, the U.S., and Singapore), our approach surpasses state-of-the-art methods across detection accuracy, geographic coverage, and cross-regional consistency. To our knowledge, this is the first method enabling end-to-end, field-collection-free generation of semantically complete, lane-level HD maps.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Planning, Routing, and Scheduling: Planning with Language Models
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

High-definition (HD) maps, particularly those containing lane-level information regarded as ground truth, are crucial for vehicle localization research. Traditionally, constructing HD maps requires highly accurate sensor measurements collection from the target area, followed by manual annotation to assign semantic information. Consequently, HD maps are limited in terms of geographic coverage. To tackle this problem, in this paper, we propose SIO-Mapper, a novel lane-level HD map construction framework that constructs city-scale maps without physical site visits by utilizing satellite images and OpenStreetmap data. One of the key contributions of SIO-Mapper is its ability to extract lane information more accurately by introducing SIO-Net, a novel deep learning network that integrates features from satellite image and OpenStreetmap using both Transformer-based and convolution-based encoders. Furthermore, to overcome challenges in merging lanes over large areas, we introduce a novel lane integration methodology that combines cluster-based and graph-based approaches. This algorithm ensures the seamless aggregation of lane segments with high accuracy and coverage, even in complex road environments. We validated SIO-Mapper on the Naver Labs Open Dataset and NuScenes dataset, demonstrating better performance in various environments including Korea, the United States, and Singapore compared to the state-of-the-art lane-level HD mapconstruction methods.

Problem

• Constructs lane-level HD maps without on-site visits
• Integrates satellite images and OpenStreetMap data
• Improves lane extraction accuracy in complex environments

Innovation

• Uses satellite images and OpenStreetMap data
• Integrates Transformer and convolution encoders
• Combines cluster and graph-based lane merging