mlsanityData to 7 Oct 2026

RoadTracer: Automatic Extraction of Road Networks from Aerial Images

arXiv 1802.03680 ↗First posted 11 Feb 2018cs.CVcited by 102

Not on the map. Why?

Abstract (arXiv)

Mapping road networks is currently both expensive and labor-intensive. High-resolution aerial imagery provides a promising avenue to automatically infer a road network. Prior work uses convolutional neural networks (CNNs) to detect which pixels belong to a road (segmentation), and then uses complex post-processing heuristics to infer graph connectivity. We show that these segmentation methods have high error rates because noisy CNN outputs are difficult to correct. We propose RoadTracer, a new method to automatically construct accurate road network maps from aerial images. RoadTracer uses an iterative search process guided by a CNN-based decision function to derive the road network graph directly from the output of the CNN. We compare our approach with a segmentation method on fifteen cities, and find that at a 5% error rate, RoadTracer correctly captures 45% more junctions across these cities.

Key facts. Cited by 102 papers in the map (data through 7 Oct 2026). First posted on arXiv 11 Feb 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 22; 2022: 16; 2023: 17; 2024: 12; 2025: 19; 2026: 16.

102 papers.

Sort:

By area:

  1. 2021MP3: A Unified Model to Map, Perceive, Predict and Plancited by 211
  2. 2022SatlasPretrain: A Large-Scale Dataset for Remote Sensing Image Understandingcited by 164
  3. 2022Relationformer: A Unified Framework for Image-to-Graph Generationcited by 67
  4. 2023Lane Graph as Path: Continuity-preserving Path-wise Modeling for Online Lane Graph Constructioncited by 50
  5. 2021Geometry-Guided Street-View Panorama Synthesis from Satellite Imagerycited by 41
  6. 2021HDMapGen: A Hierarchical Graph Generative Model of High Definition Mapscited by 40
  7. 2022EarthNets: Empowering AI in Earth Observationcited by 40
  8. 2021PointFlow: Flowing Semantics Through Points for Aerial Image Segmentationcited by 29
  9. 2022RNGDet: Road Network Graph Detection by Transformer in Aerial Imagescited by 26
  10. 2021Structure-Aware Image Segmentation with Homotopy Warpingcited by 24
  11. 2022RNGDet++: Road Network Graph Detection by Transformer with Instance Segmentation and Multi-scale Features Enhancementcited by 24
  12. 2021iCurb: Imitation Learning-based Detection of Road Curbs using Aerial Images for Autonomous Drivingcited by 21
38 more
  1. 2021Stagewise Unsupervised Domain Adaptation with Adversarial Self-Training for Road Segmentation of Remote Sensing Imagescited by 20
  2. 2023Learning and Aggregating Lane Graphs for Urban Automated Drivingcited by 20
  3. 2021Topo-boundary: A Benchmark Dataset on Topological Road-boundary Detection Using Aerial Images for Autonomous Drivingcited by 19
  4. 2023InsMapper: Exploring Inner-instance Information for Vectorized HD Mappingcited by 19
  5. 2021Lane Graph Estimation for Scene Understanding in Urban Drivingcited by 18
  6. 2021SPIN Road Mapper: Extracting Roads from Aerial Images via Spatial and Interaction Space Graph Reasoning for Autonomous Drivingcited by 18
  7. 2024SG-BEV: Satellite-Guided BEV Fusion for Cross-View Semantic Segmentationcited by 18
  8. 2022CenterLineDet: CenterLine Graph Detection for Road Lanes with Vehicle-mounted Sensors by Transformer for HD Map Generationcited by 15
  9. 2022High-Definition Map Generation Technologies For Autonomous Drivingcited by 14
  10. 2021House-GAN++: Generative Adversarial Layout Refinement Networkscited by 13
  11. 2023Beyond Geo-localization: Fine-grained Orientation of Street-view Images by Cross-view Matching with Satellite Imagery with Supplementary Materialscited by 13
  12. 2021csBoundary: City-scale Road-boundary Detection in Aerial Images for High-definition Mapscited by 12
  13. 2023Building Lane-Level Maps from Aerial Imagescited by 12
  14. 2024Enabling Uncertainty Estimation in Iterative Neural Networkscited by 11
  15. 2024OpenSatMap: A Fine-grained High-resolution Satellite Dataset for Large-scale Map Constructioncited by 11
  16. 2021Persistent Homology with Improved Locality Information for more Effective Delineationcited by 10
  17. 2022DDU-Net: Dual-Decoder-U-Net for Road Extraction Using High-Resolution Remote Sensing Imagescited by 10
  18. 2024Towards Satellite Image Road Graph Extraction: A Global-Scale Dataset and A Novel Methodcited by 10
  19. 2021CP-loss: Connectivity-preserving Loss for Road Curb Detection in Autonomous Driving with Aerial Imagescited by 9
  20. 2023VMA: Divide-and-Conquer Vectorized Map Annotation System for Large-Scale Driving Scenecited by 9
  21. 2021Graph Representation Learning for Road Type Classificationcited by 8
  22. 2024LOGCAN++: Adaptive Local-global class-aware network for semantic segmentation of remote sensing imagerycited by 7
  23. 2021Trajectory Prediction for Autonomous Driving with Topometric Mapcited by 6
  24. 2024IT3: Idempotent Test-Time Trainingcited by 6
  25. 2021Beyond Road Extraction: A Dataset for Map Update using Aerial Imagescited by 5
  26. 2023Patched Line Segment Learning for Vector Road Mappingcited by 5
  27. 2024Continuity Preserving Online CenterLine Graph Learningcited by 5
  28. 2025LaneDiffusion: Improving Centerline Graph Learning via Prior Injected BEV Feature Generationcited by 5
  29. 2021YOLinO: Generic Single Shot Polyline Detection in Real Timecited by 4
  30. 2021Hierarchical Road Topology Learning for Urban Map-less Drivingcited by 4
  31. 2022Progressive Minimal Path Method with Embedded CNNcited by 4
  32. 2023Road Extraction with Satellite Images and Partial Road Mapscited by 4
  33. 2023AutoGraph: Predicting Lane Graphs from Traffic Observationscited by 4
  34. 2025QuFeX: Quantum feature extraction module for hybrid quantum-classical deep neural networkscited by 4
  35. 2025HoliTracer: Holistic Vectorization of Geographic Objects from Large-Size Remote Sensing Imagerycited by 4
  36. 2025IRSAMap:Towards Large-Scale, High-Resolution Land Cover Map Vectorizationcited by 4
  37. 2021GridTracer: Automatic Mapping of Power Grids using Deep Learning and Overhead Imagerycited by 3
  38. 2021Updating Street Maps using Changes Detected in Satellite Imagerycited by 3

50 most cited of 102 All 102 (JSON)