# Leveraging Crowdsourced GPS Data for Road Extraction from Aerial Imagery

arXiv: https://arxiv.org/abs/1905.01447 · HTML: https://mlsanity.com/paper/1905.01447/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 May 2019.

## Built on by (most cited first, 23 of 23)

- [Structured Bird's-Eye-View Traffic Scene Understanding from Onboard Images](https://mlsanity.com/paper/2110.01997/) (2021)
- [Topology Preserving Local Road Network Estimation from Single Onboard Camera Image](https://mlsanity.com/paper/2112.10155/) (2021)
- [GLF-CR: SAR-Enhanced Cloud Removal with Global-Local Fusion](https://mlsanity.com/paper/2206.02850/) (2022)
- [Spatio-channel Attention Blocks for Cross-modal Crowd Counting](https://mlsanity.com/paper/2210.10392/) (2022)
- [Improving Online Lane Graph Extraction by Object-Lane Clustering](https://mlsanity.com/paper/2307.10947/) (2023)
- [Aerial Images Meet Crowdsourced Trajectories: A New Approach to Robust Road Extraction](https://mlsanity.com/paper/2111.15119/) (2021)
- [Beyond Road Extraction: A Dataset for Map Update using Aerial Images](https://mlsanity.com/paper/2110.04690/) (2021)
- [Monocular BEV Perception of Road Scenes via Front-to-Top View Projection](https://mlsanity.com/paper/2211.08144/) (2022)
- [Road Extraction with Satellite Images and Partial Road Maps](https://mlsanity.com/paper/2303.12394/) (2023)
- [PathwayBench: Assessing Routability of Pedestrian Pathway Networks Inferred from Multi-City Imagery](https://mlsanity.com/paper/2407.16875/) (2024)
- [Land Cover Mapping in Limited Labels Scenario: A Survey](https://mlsanity.com/paper/2103.02429/) (2021)
- [APE: An Open and Shared Annotated Dataset for Learning Urban Pedestrian Path Networks](https://mlsanity.com/paper/2303.02323/) (2023)
- [Prior Based Online Lane Graph Extraction from Single Onboard Camera Image](https://mlsanity.com/paper/2307.13344/) (2023)
- [Reliable, Routable, and Reproducible: Collection of Pedestrian Pathways at Statewide Scale](https://mlsanity.com/paper/2410.19762/) (2024)
- [EAA-Net: Rethinking the Autoencoder Architecture with Intra-class Features for Medical Image Segmentation](https://mlsanity.com/paper/2208.09197/) (2022)
- [Leveraging GNSS and Onboard Visual Data from Consumer Vehicles for Robust Road Network Estimation](https://mlsanity.com/paper/2408.01640/) (2024)
- [Semi-Supervised Domain Adaptation for Semantic Segmentation of Roads from Satellite Images](https://mlsanity.com/paper/2212.13079/) (2022)
- [MultiScale Probability Map guided Index Pooling with Attention-based learning for Road and Building Segmentation](https://mlsanity.com/paper/2302.09411/) (2023)
- [Online Lane Graph Extraction from Onboard Video](https://mlsanity.com/paper/2304.00930/) (2023)
- [Large-scale Weakly Supervised Learning for Road Extraction from Satellite Imagery](https://mlsanity.com/paper/2309.07823/) (2023)
- [AI Powered Road Network Prediction with Multi-Modal Data](https://mlsanity.com/paper/2312.17040/) (2023)
- [Performance Evaluation of GPS Trajectory Rasterization Methods](https://mlsanity.com/paper/2401.01676/) (2024)
- [TraversRL: Traversable Pedestrian Pathway Generation With Reinforcement Learning](https://mlsanity.com/paper/2607.17479/) (2026)

## Abstract (arXiv)

Deep learning is revolutionizing the mapping industry. Under lightweight human curation, computer has generated almost half of the roads in Thailand on OpenStreetMap (OSM) using high-resolution aerial imagery. Bing maps are displaying 125 million computer-generated building polygons in the U.S. While tremendously more efficient than manual mapping, one cannot map out everything from the air. Especially for roads, a small prediction gap by image occlusion renders the entire road useless for routing. Misconnections can be more dangerous. Therefore computer-based mapping often requires local verifications, which is still labor intensive. In this paper, we propose to leverage crowdsourced GPS data to improve and support road extraction from aerial imagery. Through novel data augmentation, GPS rendering, and 1D transpose convolution techniques, we show almost 5% improvements over previous competition winning models, and much better robustness when predicting new areas without any new training data or domain adaptation.