mlsanityData to 7 Oct 2026

HDMapGen: A Hierarchical Graph Generative Model of High Definition Maps

arXiv 2106.14880 ↗First posted 28 Jun 2021cs.CVcited by 40

Not on the map. Why?

Abstract (arXiv)

High Definition (HD) maps are maps with precise definitions of road lanes with rich semantics of the traffic rules. They are critical for several key stages in an autonomous driving system, including motion forecasting and planning. However, there are only a small amount of real-world road topologies and geometries, which significantly limits our ability to test out the self-driving stack to generalize onto new unseen scenarios. To address this issue, we introduce a new challenging task to generate HD maps. In this work, we explore several autoregressive models using different data representations, including sequence, plain graph, and hierarchical graph. We propose HDMapGen, a hierarchical graph generation model capable of producing high-quality and diverse HD maps through a coarse-to-fine approach. Experiments on the Argoverse dataset and an in-house dataset show that HDMapGen significantly outperforms baseline methods. Additionally, we demonstrate that HDMapGen achieves high scalability and efficiency.

Key facts. Cited by 40 papers in the map (data through 7 Oct 2026). Builds on 21 of its 40 references that are in the map. First posted on arXiv 28 Jun 2021.

Builds on

21 of 40 references.

References · 21

  1. 2017Graph Attention Networkscited by 8,854
  2. 2014Sequence to Sequence Learning with Neural Networkscited by 2,974
  3. 2016WaveNet: A Generative Model for Raw Audiocited by 1,710
  4. 2016Variational Graph Auto-Encoderscited by 1,373
  5. 2019Argoverse: 3D Tracking and Forecasting with Rich Mapscited by 728
  6. 2020VectorNet: Encoding HD Maps and Agent Dynamics from Vectorized Representationcited by 528
  7. 2019MultiPath: Multiple Probabilistic Anchor Trajectory Hypotheses for Behavior Predictioncited by 417
  8. 2018GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Modelscited by 406
  9. 2018MolGAN: An implicit generative model for small molecular graphscited by 392
  10. 2020Learning Lane Graph Representations for Motion Forecastingcited by 380
  11. 2020TNT: Target-driveN Trajectory Predictioncited by 345
  12. 2017A Neural Representation of Sketch Drawingscited by 311
9 more
  1. 2020PolyGen: An Autoregressive Generative Model of 3D Meshescited by 267
  2. 2019Efficient Graph Generation with Graph Recurrent Attention Networkscited by 231
  3. 2017GraphGAN: Graph Representation Learning with Generative Adversarial Netscited by 121
  4. 2018RoadTracer: Automatic Extraction of Road Networks from Aerial Imagescited by 102
  5. 2019Variational Graph Recurrent Neural Networkscited by 86
  6. 2018NeVAE: A Deep Generative Model for Molecular Graphscited by 66
  7. 2013Deep AutoRegressive Networkscited by 60
  8. 2020DAGMapper: Learning to Map by Discovering Lane Topologycited by 49
  9. 2019Neural Turtle Graphics for Modeling City Road Layoutscited by 29

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 3; 2022: 3; 2023: 16; 2024: 8; 2025: 6; 2026: 4.

40 papers.

Sort:

By area:

  1. 2023End-to-end Autonomous Driving: Challenges and Frontierscited by 485
  2. 2021HDMapNet: An Online HD Map Construction and Evaluation Frameworkcited by 280
  3. 2022VectorMapNet: End-to-end Vectorized HD Map Learningcited by 243
  4. 2023PivotNet: Vectorized Pivot Learning for End-to-end HD Map Constructioncited by 84
  5. 2023End-to-End Vectorized HD-map Construction with Piecewise Bezier Curvecited by 80
  6. 2023Lane Graph as Path: Continuity-preserving Path-wise Modeling for Online Lane Graph Constructioncited by 50
  7. 2024HIMap: HybrId Representation Learning for End-to-end Vectorized HD Map Constructioncited by 48
  8. 2023Online Vectorized HD Map Construction using Geometrycited by 42
  9. 2021Vehicle trajectory prediction works, but not everywherecited by 39
  10. 2023The Integration of Prediction and Planning in Deep Learning Automated Driving Systems: A Reviewcited by 29
  11. 2024Segment Anything Model for Road Network Graph Extractioncited by 28
  12. 2024SLEDGE: Synthesizing Driving Environments with Generative Models and Rule-Based Trafficcited by 27
28 more
  1. 2025Scenario Dreamer: Vectorized Latent Diffusion for Generating Driving Simulation Environmentscited by 27
  2. 2023DriveSceneGen: Generating Diverse and Realistic Driving Scenarios from Scratchcited by 26
  3. 2023Coarse-to-Fine: a Hierarchical Diffusion Model for Molecule Generation in 3Dcited by 24
  4. 2023InsMapper: Exploring Inner-instance Information for Vectorized HD Mappingcited by 19
  5. 2021csBoundary: City-scale Road-boundary Detection in Aerial Images for High-definition Mapscited by 12
  6. 2024DTCLMapper: Dual Temporal Consistent Learning for Vectorized HD Map Constructioncited by 10
  7. 2023End-to-end Autonomous Driving using Deep Learning: A Systematic Reviewcited by 7
  8. 2023Pix2Map: Cross-modal Retrieval for Inferring Street Maps from Imagescited by 6
  9. 2025AugMapNet: Improving Spatial Latent Structure via BEV Grid Augmentation for Enhanced Vectorized Online HD Map Constructioncited by 6
  10. 2023Exploring Navigation Maps for Learning-Based Motion Predictioncited by 5
  11. 2024PathwayBench: Assessing Routability of Pedestrian Pathway Networks Inferred from Multi-City Imagerycited by 4
  12. 2024Neural HD Map Generation from Multiple Vectorized Tiles Locally Produced by Autonomous Vehiclescited by 4
  13. 2023APE: An Open and Shared Annotated Dataset for Learning Urban Pedestrian Path Networkscited by 3
  14. 2023Learning to Predict Navigational Patterns from Partial Observationscited by 3
  15. 2025Generative AI for Autonomous Driving: A Reviewcited by 3
  16. 2022FLOWGEN: Fast and slow graph generationcited by 2
  17. 2022Large-Scale Auto-Regressive Modeling Of Street Networkscited by 2
  18. 2023OASIS: Automated Assessment of Urban Pedestrian Paths at Scalecited by 2
  19. 2024LMT-Net: Lane Model Transformer Network for Automated HD Mapping from Sparse Vehicle Observationscited by 2
  20. 2024Reliable, Routable, and Reproducible: Collection of Pedestrian Pathways at Statewide Scalecited by 2
  21. 2025A Concise Survey on Lane Topology Reasoning for HD Mappingcited by 2
  22. 2023Procedural Generation of Complex Roundabouts for Autonomous Vehicle Testingcited by 1
  23. 2025PrITTI: Primitive-based Generation of Controllable and Editable 3D Semantic Urban Scenescited by 1
  24. 2025Maps for Autonomous Driving: Full-process Survey and Frontierscited by 1
  25. 2026ControlMap: Controllable High-Definition Map Generation for Traffic Scenario Simulationcited by 1
  26. 2026ScenarioControl: Vision-Language Controllable Vectorized Latent Scenario Generationcited by 0
  27. 2026TraversRL: Traversable Pedestrian Pathway Generation With Reinforcement Learningcited by 0
  28. 2026RoadWeaver: Large-Scale Lane-Level HD Map Generation from Scratch for Autonomous Driving Simulationcited by 0