mlsanityData to 7 Oct 2026

RoadBEV: Road Surface Reconstruction in Bird's Eye View

arXiv 2404.06605 ↗First posted 9 Apr 2024cs.CVcited by 8

Not on the map. Why?

Abstract (arXiv)

Road surface conditions, especially geometry profiles, enormously affect driving performance of autonomous vehicles. Vision-based online road reconstruction promisingly captures road information in advance. Existing solutions like monocular depth estimation and stereo matching suffer from modest performance. The recent technique of Bird's-Eye-View (BEV) perception provides immense potential to more reliable and accurate reconstruction. This paper uniformly proposes two simple yet effective models for road elevation reconstruction in BEV named RoadBEV-mono and RoadBEV-stereo, which estimate road elevation with monocular and stereo images, respectively. The former directly fits elevation values based on voxel features queried from image view, while the latter efficiently recognizes road elevation patterns based on BEV volume representing correlation between left and right voxel features. Insightful analyses reveal their consistence and difference with the perspective view. Experiments on real-world dataset verify the models' effectiveness and superiority. Elevation errors of RoadBEV-mono and RoadBEV-stereo achieve 1.83 cm and 0.50 cm, respectively. Our models are promising for practical road preview, providing essential information for promoting safety and comfort of autonomous vehicles. The code is released at https://github.com/ztsrxh/RoadBEV

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 40 of its 71 references that are in the map. First posted on arXiv 9 Apr 2024.

Builds on

40 of 71 references.

References · 40

  1. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
  2. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,141
  3. 2022BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformerscited by 1,213
  4. 2020Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3Dcited by 1,040
  5. 2021BEVDet: High-performance Multi-camera 3D Object Detection in Bird-Eye-Viewcited by 711
  6. 2020AdaBins: Depth Estimation using Adaptive Binscited by 596
  7. 2022BEVDepth: Acquisition of Reliable Depth for Multi-view 3D Object Detectioncited by 568
  8. 2023DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-based 3D Visioncited by 514
  9. 2018Pyramid Stereo Matching Networkcited by 486
  10. 2023Tri-Perspective View for Vision-Based 3D Semantic Occupancy Predictioncited by 404
  11. 2022BEVDet4D: Exploit Temporal Cues in Multi-camera 3D Object Detectioncited by 340
  12. 2021Categorical Depth Distribution Network for Monocular 3D Object Detectioncited by 326
28 more
  1. 2021HDMapNet: An Online HD Map Construction and Evaluation Frameworkcited by 280
  2. 2019Group-wise Correlation Stereo Networkcited by 235
  3. 2023Scene as Occupancycited by 208
  4. 2023Iterative Geometry Encoding Volume for Stereo Matchingcited by 188
  5. 2022M2BEV: Multi-Camera Joint 3D Detection and Segmentation with Unified Birds-Eye View Representationcited by 155
  6. 2021CFNet: Cascade and Fused Cost Volume for Robust Stereo Matchingcited by 153
  7. 2022Attention Concatenation Volume for Accurate and Efficient Stereo Matchingcited by 130
  8. 2022Delving into the Devils of Bird's-eye-view Perception: A Review, Evaluation and Recipecited by 122
  9. 2022Attention Attention Everywhere: Monocular Depth Prediction with Skip Attentioncited by 92
  10. 2022BEVStereo: Enhancing Depth Estimation in Multi-view 3D Object Detection with Dynamic Temporal Stereocited by 89
  11. 2023BEVHeight: A Robust Framework for Vision-based Roadside 3D Object Detectioncited by 74
  12. 2023iDisc: Internal Discretization for Monocular Depth Estimationcited by 71
  13. 2023VMT-Adapter: Parameter-Efficient Transfer Learning for Multi-Task Dense Scene Understandingcited by 62
  14. 2024GraphBEV: Towards Robust BEV Feature Alignment for Multi-Modal 3D Object Detectioncited by 33
  15. 2023Dual Radar: A Multi-modal Dataset with Dual 4D Radar for Autonomous Drivingcited by 32
  16. 2023BEVHeight++: Toward Robust Visual Centric 3D Object Detectioncited by 29
  17. 2024DDN-SLAM: Real-time Dense Dynamic Neural Implicit SLAMcited by 22
  18. 2024TaskCLIP: Extend Large Vision-Language Model for Task Oriented Object Detectioncited by 21
  19. 2023RoMe: Towards Large Scale Road Surface Reconstruction via Mesh Representationcited by 20
  20. 2023MonoGAE: Roadside Monocular 3D Object Detection with Ground-Aware Embeddingscited by 20
  21. 2024RoboFusion: Towards Robust Multi-Modal 3D Object Detection via SAMcited by 20
  22. 2020BirdNet+: End-to-End 3D Object Detection in LiDAR Bird's Eye Viewcited by 19
  23. 2022Efficient Game-Theoretic Planning with Prediction Heuristic for Socially-Compliant Autonomous Drivingcited by 13
  24. 2024EMIE-MAP: Large-Scale Road Surface Reconstruction Based on Explicit Mesh and Implicit Encodingcited by 13
  25. 2023RSRD: A Road Surface Reconstruction Dataset and Benchmark for Safe and Comfortable Autonomous Drivingcited by 8
  26. 2019Real-Time Dense Stereo Embedded in A UAV for Road Inspectioncited by 7
  27. 2021Pseudo-LiDAR Based Road Detectioncited by 6
  28. 2024Depth-aware Volume Attention for Texture-less Stereo Matchingcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 5; 2025: 2; 2026: 1.

8 papers.

  1. 2024V2X-Radar: A Multi-modal Dataset with 4D Radar for Cooperative Perceptioncited by 21
  2. 2024Event-guided Low-light Video Semantic Segmentationcited by 14
  3. 2025Collaborative Perception Datasets for Autonomous Driving: A Reviewcited by 8
  4. 2024RoGs: Large Scale Road Surface Reconstruction with Meshgrid Gaussiancited by 5
  5. 2024MonoPP: Metric-Scaled Self-Supervised Monocular Depth Estimation by Planar-Parallax Geometry in Automotive Applicationscited by 4
  6. 2025Gamma-from-Mono: Road-Relative, Metric, Self-Supervised Monocular Geometry for Vehicular Applicationscited by 1
  7. 2024Establishing Reality-Virtuality Interconnections in Urban Digital Twins for Superior Intelligent Road Inspection and Simulationcited by 0
  8. 2026RoGS: Adaptive Meshgrid Gaussian for Large-Scale Road Surface Mappingcited by 0