mlsanityData to 14 Oct 2026

SafeMap: Robust HD Map Construction from Incomplete Observations

arXiv 2507.00861 ↗First posted 1 Jul 2025cs.CVcited by 14

Not on the map. Why?

Abstract (arXiv)

Robust high-definition (HD) map construction is vital for autonomous driving, yet existing methods often struggle with incomplete multi-view camera data. This paper presents SafeMap, a novel framework specifically designed to secure accuracy even when certain camera views are missing. SafeMap integrates two key components: the Gaussian-based Perspective View Reconstruction (G-PVR) module and the Distillation-based Bird's-Eye-View (BEV) Correction (D-BEVC) module. G-PVR leverages prior knowledge of view importance to dynamically prioritize the most informative regions based on the relationships among available camera views. Furthermore, D-BEVC utilizes panoramic BEV features to correct the BEV representations derived from incomplete observations. Together, these components facilitate the end-to-end map reconstruction and robust HD map generation. SafeMap is easy to implement and integrates seamlessly into existing systems, offering a plug-and-play solution for enhanced robustness. Experimental results demonstrate that SafeMap significantly outperforms previous methods in both complete and incomplete scenarios, highlighting its superior performance and reliability.

Key facts. Cited by 14 papers in the map (data through 14 Oct 2026). Builds on 54 of its 63 references that are in the map. First posted on arXiv 1 Jul 2025.

Builds on

54 of 63 references.

References · 54

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,523
  3. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,422
  4. 2022BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformerscited by 1,218
  5. 2022BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representationcited by 850
  6. 2023Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecastingcited by 707
  7. 2023End-to-end Autonomous Driving: Challenges and Frontierscited by 490
  8. 2020Calibrating Deep Neural Networks using Focal Losscited by 357
  9. 2023Repurposing Diffusion-Based Image Generators for Monocular Depth Estimationcited by 327
  10. 2023SurroundOcc: Multi-Camera 3D Occupancy Prediction for Autonomous Drivingcited by 316
  11. 2022MapTR: Structured Modeling and Learning for Online Vectorized HD Map Constructioncited by 314
  12. 2021HDMapNet: An Online HD Map Construction and Evaluation Frameworkcited by 281
38 more
  1. 2022Cross-view Transformers for real-time Map-view Semantic Segmentationcited by 264
  2. 2022VectorMapNet: End-to-end Vectorized HD Map Learningcited by 245
  3. 2023MapTRv2: An End-to-End Framework for Online Vectorized HD Map Constructioncited by 193
  4. 2022BEVerse: Unified Perception and Prediction in Birds-Eye-View for Vision-Centric Autonomous Drivingcited by 191
  5. 2023StreamMapNet: Streaming Mapping Network for Vectorized Online HD Map Constructioncited by 145
  6. 2023Robo3D: Towards Robust and Reliable 3D Perception against Corruptionscited by 129
  7. 2023Rethinking Range View Representation for LiDAR Segmentationcited by 128
  8. 2022Vision-Centric BEV Perception: A Surveycited by 126
  9. 2023Segment Any Point Cloud Sequences by Distilling Vision Foundation Modelscited by 125
  10. 2023Benchmarking Robustness of 3D Object Detection to Common Corruptions in Autonomous Drivingcited by 114
  11. 2024SparseOcc: Rethinking Sparse Latent Representation for Vision-Based Semantic Occupancy Predictioncited by 96
  12. 2023PivotNet: Vectorized Pivot Learning for End-to-end HD Map Constructioncited by 85
  13. 2023End-to-End Vectorized HD-map Construction with Piecewise Bezier Curvecited by 82
  14. 2022Efficient and Robust 2D-to-BEV Representation Learning via Geometry-guided Kernel Transformercited by 80
  15. 2022SurroundDepth: Entangling Surrounding Views for Self-Supervised Multi-Camera Depth Estimationcited by 76
  16. 2025MapNav: A Novel Memory Representation via Annotated Semantic Maps for Vision-and-Language Navigationcited by 60
  17. 2024IS-Fusion: Instance-Scene Collaborative Fusion for Multimodal 3D Object Detectioncited by 57
  18. 2023Online Map Vectorization for Autonomous Driving: A Rasterization Perspectivecited by 54
  19. 2024Multi-Modal Data-Efficient 3D Scene Understanding for Autonomous Drivingcited by 54
  20. 2024MGMap: Mask-Guided Learning for Online Vectorized HD Map Constructioncited by 52
  21. 2023FRNet: Frustum-Range Networks for Scalable LiDAR Segmentationcited by 50
  22. 2023RoboDepth: Robust Out-of-Distribution Depth Estimation under Corruptionscited by 49
  23. 2024HIMap: HybrId Representation Learning for End-to-end Vectorized HD Map Constructioncited by 48
  24. 2023RoboBEV: Towards Robust Bird's Eye View Perception under Corruptionscited by 42
  25. 2024Benchmarking and Improving Bird's Eye View Perception Robustness in Autonomous Drivingcited by 41
  26. 2022SuperFusion: Multilevel LiDAR-Camera Fusion for Long-Range HD Map Generationcited by 39
  27. 2023Understanding the Robustness of 3D Object Detection with Bird's-Eye-View Representations in Autonomous Drivingcited by 37
  28. 2024GraphBEV: Towards Robust BEV Feature Alignment for Multi-Modal 3D Object Detectioncited by 33
  29. 20244D Contrastive Superflows are Dense 3D Representation Learnerscited by 33
  30. 2023UniBEV: Multi-modal 3D Object Detection with Uniform BEV Encoders for Robustness against Missing Sensor Modalitiescited by 29
  31. 2024Enhancing Vectorized Map Perception with Historical Rasterized Mapscited by 29
  32. 2024Is Your HD Map Constructor Reliable under Sensor Corruptions?cited by 28
  33. 2024Is Your LiDAR Placement Optimized for 3D Scene Understanding?cited by 26
  34. 2024MapDistill: Boosting Efficient Camera-based HD Map Construction via Camera-LiDAR Fusion Model Distillationcited by 26
  35. 2024Multi-Space Alignments Towards Universal LiDAR Segmentationcited by 25
  36. 2024The RoboDrive Challenge: Drive Anytime Anywhere in Any Conditioncited by 23
  37. 2025MapFusion: A Novel BEV Feature Fusion Network for Multi-modal Map Constructioncited by 21
  38. 2023M-BEV: Masked BEV Perception for Robust Autonomous Drivingcited by 19

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 8; 2026: 6.

14 papers.

Sort:

By area:

  1. 2025Vision-Language-Action Models for Autonomous Driving: Past, Present, and Futurecited by 33
  2. 2025GuideFlow: Constraint-Guided Flow Matching for Planning in End-to-End Autonomous Drivingcited by 22
  3. 2025What Really Matters for Robust Multi-Sensor HD Map Construction?cited by 9
  4. 2025Stability Under Scrutiny: Benchmarking Representation Paradigms for Online HD Mappingcited by 7
  5. 2025MoRA: Missing Modality Low-Rank Adaptation for Visual Recognitioncited by 4
  6. 2025DSBench: A Comprehensive Benchmark for Evaluating External and In-Cabin Riskscited by 3
  7. 2025AMap: Distilling Future Priors for Ahead-Aware Online HD Map Constructioncited by 2
  8. 2026FlexMap: Robust HD Map Construction under Flexible Camera Configurationscited by 2
  9. 2026SEF-MAP: Subspace-Decomposed Expert Fusion for Robust Multimodal HD Map Predictioncited by 2
  10. 2026M2-Occ: Resilient 3D Semantic Occupancy Prediction for Autonomous Driving with Incomplete Camera Inputscited by 1
  11. 2026Semantic-Aware, Physics-Informed, Geometry-Grounded Weather Video Synthesiscited by 1
  12. 2025Learning to Navigate Socially Through Proactive Risk Perceptioncited by 0
2 more
  1. 2026SpikeOcc: Towards Resilient Multimodal 3D Occupancy Perception via Spiking Expertscited by 0
  2. 2026The Value of Being Wrong: Self-Mined Visual In-Context Learning from Errorscited by 0