mlsanityData to 13 Oct 2026

SparseBEV: High-Performance Sparse 3D Object Detection from Multi-Camera Videos

arXiv 2308.09244 ↗First posted 18 Aug 2023cs.CVcited by 150

Not on the map. Why?

Abstract (arXiv)

Camera-based 3D object detection in BEV (Bird's Eye View) space has drawn great attention over the past few years. Dense detectors typically follow a two-stage pipeline by first constructing a dense BEV feature and then performing object detection in BEV space, which suffers from complex view transformations and high computation cost. On the other side, sparse detectors follow a query-based paradigm without explicit dense BEV feature construction, but achieve worse performance than the dense counterparts. In this paper, we find that the key to mitigate this performance gap is the adaptability of the detector in both BEV and image space. To achieve this goal, we propose SparseBEV, a fully sparse 3D object detector that outperforms the dense counterparts. SparseBEV contains three key designs, which are (1) scale-adaptive self attention to aggregate features with adaptive receptive field in BEV space, (2) adaptive spatio-temporal sampling to generate sampling locations under the guidance of queries, and (3) adaptive mixing to decode the sampled features with dynamic weights from the queries. On the test split of nuScenes, SparseBEV achieves the state-of-the-art performance of 67.5 NDS. On the val split, SparseBEV achieves 55.8 NDS while maintaining a real-time inference speed of 23.5 FPS. Code is available at https://github.com/MCG-NJU/SparseBEV.

Key facts. Cited by 150 papers in the map (data through 13 Oct 2026). Builds on 54 of its 65 references that are in the map. First posted on arXiv 18 Aug 2023.

Builds on

54 of 65 references.

References · 54

  1. 2017Attention Is All You Needcited by 60,106
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,950
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,043
  4. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,223
  5. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,111
  6. 2020End-to-End Object Detection with Transformerscited by 7,396
  7. 2017Mask R-CNNcited by 7,266
  8. 2016Feature Pyramid Networks for Object Detectioncited by 5,027
  9. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,498
  10. 2022A ConvNet for the 2020scited by 3,563
  11. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,417
  12. 2018SlowFast Networks for Video Recognitioncited by 1,864
38 more
  1. 2018PointPillars: Fast Encoders for Object Detection from Point Cloudscited by 1,709
  2. 2019FCOS: Fully Convolutional One-Stage Object Detectioncited by 1,575
  3. 2021MLP-Mixer: An all-MLP Architecture for Visioncited by 1,442
  4. 2022DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detectioncited by 1,312
  5. 2022BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformerscited by 1,215
  6. 2020Center-based 3D Object Detection and Trackingcited by 1,144
  7. 2020Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3Dcited by 1,040
  8. 2017Focal Loss for Dense Object Detectioncited by 750
  9. 2021BEVDet: High-performance Multi-camera 3D Object Detection in Bird-Eye-Viewcited by 711
  10. 2021DETR3D: 3D Object Detection from Multi-view Images via 3D-to-2D Queriescited by 658
  11. 2022DAB-DETR: Dynamic Anchor Boxes are Better Queries for DETRcited by 571
  12. 2022BEVDepth: Acquisition of Reliable Depth for Multi-view 3D Object Detectioncited by 569
  13. 2022PETR: Position Embedding Transformation for Multi-View 3D Object Detectioncited by 551
  14. 2020Sparse R-CNN: End-to-End Object Detection with Learnable Proposalscited by 495
  15. 2022DN-DETR: Accelerate DETR Training by Introducing Query DeNoisingcited by 488
  16. 2022InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutionscited by 482
  17. 2021FCOS3D: Fully Convolutional One-Stage Monocular 3D Object Detectioncited by 446
  18. 2021Conditional DETR for Fast Training Convergencecited by 423
  19. 2022PETRv2: A Unified Framework for 3D Perception from Multi-Camera Imagescited by 352
  20. 2019Class-balanced Grouping and Sampling for Point Cloud 3D Object Detectioncited by 350
  21. 2022BEVDet4D: Exploit Temporal Cues in Multi-camera 3D Object Detectioncited by 341
  22. 2021Categorical Depth Distribution Network for Monocular 3D Object Detectioncited by 326
  23. 2016Dynamic Filter Networkscited by 318
  24. 2023Exploring Object-Centric Temporal Modeling for Efficient Multi-View 3D Object Detectioncited by 314
  25. 2022BEVFormer v2: Adapting Modern Image Backbones to Bird's-Eye-View Recognition via Perspective Supervisioncited by 302
  26. 2022SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentationcited by 252
  27. 2021Is Pseudo-Lidar needed for Monocular 3D Object detection?cited by 248
  28. 2022Unifying Voxel-based Representation with Transformer for 3D Object Detectioncited by 202
  29. 2018Orthographic Feature Transform for Monocular 3D Object Detectioncited by 201
  30. 2019CenterMask : Real-Time Anchor-Free Instance Segmentationcited by 200
  31. 2022Time Will Tell: New Outlooks and A Baseline for Temporal Multi-View 3D Object Detectioncited by 178
  32. 2019Cross-view Semantic Segmentation for Sensing Surroundingscited by 173
  33. 2022Sparse4D: Multi-view 3D Object Detection with Sparse Spatial-Temporal Fusioncited by 162
  34. 2022DETRs with Hybrid Matchingcited by 150
  35. 2022PolarFormer: Multi-camera 3D Object Detection with Polar Transformercited by 147
  36. 2023Sparse4D v2: Recurrent Temporal Fusion with Sparse Modelcited by 93
  37. 2022BEVStereo: Enhancing Depth Estimation in Multi-view 3D Object Detection with Dynamic Temporal Stereocited by 90
  38. 2022AdaMixer: A Fast-Converging Query-Based Object Detectorcited by 79

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2023: 9; 2024: 55; 2025: 52; 2026: 34.

150 papers.

Sort:

By area:

  1. 2023Is Ego Status All You Need for Open-Loop End-to-End Autonomous Driving?cited by 242
  2. 2023Visual Point Cloud Forecasting enables Scalable Autonomous Drivingcited by 97
  3. 2024RCBEVDet: Radar-camera Fusion in Bird's Eye View for 3D Object Detectioncited by 82
  4. 2023UniPAD: A Universal Pre-training Paradigm for Autonomous Drivingcited by 73
  5. 2024Driving in the Occupancy World: Vision-Centric 4D Occupancy Forecasting and Planning via World Models for Autonomous Drivingcited by 67
  6. 20253D-LLaVA: Towards Generalist 3D LMMs with Omni Superpoint Transformercited by 67
  7. 2023Fully Sparse 3D Occupancy Predictioncited by 63
  8. 2023PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigmcited by 56
  9. 2024OPUS: Occupancy Prediction Using a Sparse Setcited by 56
  10. 2024Navigation-Guided Sparse Scene Representation for End-to-End Autonomous Drivingcited by 46
  11. 2025Multimodal Fusion and Vision-Language Models: A Survey for Robot Visioncited by 38
  12. 2024A Survey on Occupancy Perception for Autonomous Driving: The Information Fusion Perspectivecited by 37
38 more
  1. 2024Ray Denoising: Depth-aware Hard Negative Sampling for Multi-view 3D Object Detectioncited by 36
  2. 2023BEVNeXt: Reviving Dense BEV Frameworks for 3D Object Detectioncited by 35
  3. 2024Unleashing HyDRa: Hybrid Fusion, Depth Consistency and Radar for Unified 3D Perceptioncited by 35
  4. 2025HiP-AD: Hierarchical and Multi-Granularity Planning with Deformable Attention for Autonomous Driving in a Single Decodercited by 35
  5. 2024OmniHD-Scenes: A Next-Generation Multimodal Dataset for Autonomous Drivingcited by 27
  6. 2024SparseLIF: High-Performance Sparse LiDAR-Camera Fusion for 3D Object Detectioncited by 25
  7. 2025Flow Matching-Based Autonomous Driving Planning with Advanced Interactive Behavior Modelingcited by 22
  8. 2025Doracamom: Joint 3D Detection and Occupancy Prediction with Multi-view 4D Radars and Cameras for Omnidirectional Perceptioncited by 21
  9. 2025STCOcc: Sparse Spatial-Temporal Cascade Renovation for 3D Occupancy and Scene Flow Predictioncited by 21
  10. 2024Is a 3D-Tokenized LLM the Key to Reliable Autonomous Driving?cited by 20
  11. 2024GaussianBeV: 3D Gaussian Representation meets Perception Models for BeV Segmentationcited by 20
  12. 2024RaCFormer: Towards High-Quality 3D Object Detection via Query-based Radar-Camera Fusioncited by 19
  13. 2024CRT-Fusion: Camera, Radar, Temporal Fusion Using Motion Information for 3D Object Detectioncited by 18
  14. 2026UniDriveVLA: Unifying Understanding, Perception, and Action Planning for Autonomous Drivingcited by 17
  15. 2024HENet: Hybrid Encoding for End-to-end Multi-task 3D Perception from Multi-view Camerascited by 16
  16. 2024UAV3D: A Large-scale 3D Perception Benchmark for Unmanned Aerial Vehiclescited by 16
  17. 2024DeepInteraction++: Multi-Modality Interaction for Autonomous Drivingcited by 15
  18. 2024DAOcc: 3D Object Detection Assisted Multi-Sensor Fusion for 3D Occupancy Predictioncited by 15
  19. 2024Enhancing 3D Object Detection with 2D Detection-Guided Query Anchorscited by 12
  20. 2024SeaBird: Segmentation in Bird's View with Dice Loss Improves Monocular 3D Detection of Large Objectscited by 12
  21. 2024CountFormer: Multi-View Crowd Counting Transformercited by 12
  22. 2024RCBEVDet++: Toward High-accuracy Radar-Camera Fusion 3D Perception Networkcited by 11
  23. 2025DiVE: Efficient Multi-View Driving Scenes Generation Based on Video Diffusion Transformercited by 10
  24. 2025SparseWorld: A Flexible, Adaptive, and Efficient 4D Occupancy World Model Powered by Sparse and Dynamic Queriescited by 10
  25. 2024DualBEV: Unifying Dual View Transformation with Probabilistic Correspondencescited by 9
  26. 2024SimPB: A Single Model for 2D and 3D Object Detection from Multiple Camerascited by 9
  27. 2024RoScenes: A Large-scale Multi-view 3D Dataset for Roadside Perceptioncited by 9
  28. 2024RayFormer: Improving Query-Based Multi-Camera 3D Object Detection via Ray-Centric Strategiescited by 9
  29. 2024Make Your ViT-based Multi-view 3D Detectors Faster via Token Compressioncited by 9
  30. 2024OpenAD: Open-World Autonomous Driving Benchmark for 3D Object Detectioncited by 9
  31. 2025RICCARDO: Radar Hit Prediction and Convolution for Camera-Radar 3D Object Detectioncited by 9
  32. 2025Bridging Perspectives: Foundation Model Guided BEV Maps for 3D Object Detection and Trackingcited by 9
  33. 2024SparseFusion: Efficient Sparse Multi-Modal Fusion Framework for Long-Range 3D Perceptioncited by 8
  34. 2024UCDNet: Multi-UAV Collaborative 3D Object Detection Network by Reliable Feature Mappingcited by 8
  35. 2024SpaRC: Sparse Radar-Camera Fusion for 3D Object Detectioncited by 8
  36. 2025I2-World: Intra-Inter Tokenization for Efficient Dynamic 4D Scene Forecastingcited by 8
  37. 2025Self-Supervised Sparse Sensor Fusion for Long Range Perceptioncited by 8
  38. 2024UniVision: A Unified Framework for Vision-Centric 3D Perceptioncited by 7

50 most cited of 150 All 150 (JSON)