mlsanityData to 14 Oct 2026

STS: Surround-view Temporal Stereo for Multi-view 3D Detection

arXiv 2208.10145 ↗First posted 22 Aug 2022cs.CVcited by 53

Not on the map. Why?

Abstract (arXiv)

Learning accurate depth is essential to multi-view 3D object detection. Recent approaches mainly learn depth from monocular images, which confront inherent difficulties due to the ill-posed nature of monocular depth learning. Instead of using a sole monocular depth method, in this work, we propose a novel Surround-view Temporal Stereo (STS) technique that leverages the geometry correspondence between frames across time to facilitate accurate depth learning. Specifically, we regard the field of views from all cameras around the ego vehicle as a unified view, namely surroundview, and conduct temporal stereo matching on it. The resulting geometrical correspondence between different frames from STS is utilized and combined with the monocular depth to yield final depth prediction. Comprehensive experiments on nuScenes show that STS greatly boosts 3D detection ability, notably for medium and long distance objects. On BEVDepth with ResNet-50 backbone, STS improves mAP and NDS by 2.6% and 1.4%, respectively. Consistent improvements are observed when using a larger backbone and a larger image resolution, demonstrating its effectiveness

Key facts. Cited by 53 papers in the map (data through 14 Oct 2026). Builds on 34 of its 38 references that are in the map. First posted on arXiv 22 Aug 2022.

Builds on

34 of 38 references.

References · 34

  1. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,523
  2. 2022BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformerscited by 1,218
  3. 2020Center-based 3D Object Detection and Trackingcited by 1,144
  4. 2020Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3Dcited by 1,044
  5. 2018MVSNet: Depth Inference for Unstructured Multi-view Stereocited by 788
  6. 2018Deep Ordinal Regression Network for Monocular Depth Estimationcited by 723
  7. 2021BEVDet: High-performance Multi-camera 3D Object Detection in Bird-Eye-Viewcited by 713
  8. 2021DETR3D: 3D Object Detection from Multi-view Images via 3D-to-2D Queriescited by 658
  9. 2020AdaBins: Depth Estimation using Adaptive Binscited by 601
  10. 2022BEVDepth: Acquisition of Reliable Depth for Multi-view 3D Object Detectioncited by 572
  11. 2022PETR: Position Embedding Transformation for Multi-View 3D Object Detectioncited by 552
  12. 2018Pyramid Stereo Matching Networkcited by 489
22 more
  1. 2021FCOS3D: Fully Convolutional One-Stage Monocular 3D Object Detectioncited by 447
  2. 2019Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matchingcited by 417
  3. 2017End-to-End Learning of Geometry and Context for Deep Stereo Regressioncited by 360
  4. 2019Class-balanced Grouping and Sampling for Point Cloud 3D Object Detectioncited by 350
  5. 2022BEVDet4D: Exploit Temporal Cues in Multi-camera 3D Object Detectioncited by 341
  6. 2019Recurrent MVSNet for High-resolution Multi-view Stereo Depth Inferencecited by 254
  7. 2019GA-Net: Guided Aggregation Net for End-to-end Stereo Matchingcited by 244
  8. 2019Group-wise Correlation Stereo Networkcited by 235
  9. 2020PatchmatchNet: Learned Multi-View Patchmatch Stereocited by 207
  10. 2020AANet: Adaptive Aggregation Network for Efficient Stereo Matchingcited by 202
  11. 2021The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depthcited by 190
  12. 2022Attention Concatenation Volume for Accurate and Efficient Stereo Matchingcited by 130
  13. 2019How do neural networks see depth in single images?cited by 82
  14. 2022Rethinking Depth Estimation for Multi-View Stereo: A Unified Representationcited by 82
  15. 2021AA-RMVSNet: Adaptive Aggregation Recurrent Multi-view Stereo Networkcited by 71
  16. 2022Multi-Frame Self-Supervised Depth with Transformerscited by 64
  17. 2019Learning Inverse Depth Regression for Multi-View Stereo with Correlation Cost Volumecited by 59
  18. 2020MonoRec: Semi-Supervised Dense Reconstruction in Dynamic Environments from a Single Moving Cameracited by 59
  19. 2021Multi-View Depth Estimation by Fusing Single-View Depth Probability with Multi-View Geometrycited by 46
  20. 2022Monocular 3D Object Detection with Depth from Motioncited by 38
  21. 2022MonoGround: Detecting Monocular 3D Objects from the Groundcited by 37
  22. 2021Depth from Camera Motion and Object Detectioncited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 8; 2023: 20; 2024: 18; 2025: 6; 2026: 1.

53 papers.

Sort:

By area:

  1. 2022BEVFormer v2: Adapting Modern Image Backbones to Bird's-Eye-View Recognition via Perspective Supervisioncited by 302
  2. 2022Time Will Tell: New Outlooks and A Baseline for Temporal Multi-View 3D Object Detectioncited by 178
  3. 2024Is Sora a World Simulator? A Comprehensive Survey on General World Models and Beyondcited by 111
  4. 2023FB-BEV: BEV Representation from Forward-Backward View Transformationscited by 98
  5. 2024DriveWorld: 4D Pre-trained Scene Understanding via World Models for Autonomous Drivingcited by 93
  6. 2022BEVPoolv2: A Cutting-edge Implementation of BEVDet Toward Deploymentcited by 79
  7. 2023BEVHeight: A Robust Framework for Vision-based Roadside 3D Object Detectioncited by 75
  8. 2023Exploring Recurrent Long-term Temporal Fusion for Multi-view 3D Perceptioncited by 65
  9. 2023Temporal Enhanced Training of Multi-view 3D Object Detector via Historical Object Predictioncited by 49
  10. 2023UniWorld: Autonomous Driving Pre-training via World Modelscited by 33
  11. 2023BEVHeight++: Toward Robust Visual Centric 3D Object Detectioncited by 30
  12. 2022MatrixVT: Efficient Multi-Camera to BEV Transformation for 3D Perceptioncited by 29
38 more
  1. 2023X3KD: Knowledge Distillation Across Modalities, Tasks and Stages for Multi-Camera 3D Object Detectioncited by 28
  2. 2022Are We Ready for Vision-Centric Driving Streaming Perception? The ASAP Benchmarkcited by 26
  3. 2023Grid-Centric Traffic Scenario Perception for Autonomous Driving: A Comprehensive Reviewcited by 25
  4. 2023Echoes Beyond Points: Unleashing the Power of Raw Radar Data in Multi-modality Fusioncited by 25
  5. 2023FrustumFormer: Adaptive Instance-aware Resampling for Multi-view 3D Detectioncited by 21
  6. 2022BEV-SAN: Accurate BEV 3D Object Detection via Slice Attention Networkscited by 20
  7. 2024CRT-Fusion: Camera, Radar, Temporal Fusion Using Motion Information for 3D Object Detectioncited by 18
  8. 2024MultiCorrupt: A Multi-Modal Robustness Dataset and Benchmark of LiDAR-Camera Fusion for 3D Object Detectioncited by 17
  9. 2024OccFeat: Self-supervised Occupancy Feature Prediction for Pretraining BEV Segmentation Networkscited by 17
  10. 2024HENet: Hybrid Encoding for End-to-end Multi-task 3D Perception from Multi-view Camerascited by 16
  11. 20223DPPE: 3D Point Positional Encoding for Multi-Camera 3D Object Detection Transformerscited by 15
  12. 2023UniScene: Multi-Camera Unified Pre-training via 3D Scene Reconstruction for Autonomous Drivingcited by 15
  13. 2023Leveraging Vision-Centric Multi-Modal Expertise for 3D Object Detectioncited by 13
  14. 2024SeaBird: Segmentation in Bird's View with Dice Loss Improves Monocular 3D Detection of Large Objectscited by 12
  15. 2024GeoBEV: Learning Geometric BEV Representation for Multi-view 3D Object Detectioncited by 11
  16. 2023SimDistill: Simulated Multi-modal Distillation for BEV 3D Object Detectioncited by 10
  17. 2023Geometric-aware Pretraining for Vision-centric 3D Object Detectioncited by 8
  18. 20233D Video Object Detection with Learnable Object-Centric Global Optimizationcited by 7
  19. 2024Towards Cross-View-Consistent Self-Supervised Surround Depth Estimationcited by 7
  20. 2024TopoMaskV2: Enhanced Instance-Mask-Based Formulation for the Road Topology Problemcited by 7
  21. 2024Autonomous Driving in Unstructured Environments: How Far Have We Come?cited by 7
  22. 2023Predict to Detect: Prediction-guided 3D Object Detection using Sequential Imagescited by 6
  23. 2024FSD-BEV: Foreground Self-Distillation for Multi-view 3D Object Detectioncited by 6
  24. 2025CHARM3R: Towards Unseen Camera Height Robust Monocular 3D Detectorcited by 6
  25. 2025HENet++: Hybrid Encoding and Multi-task Learning for 3D Perception and End-to-end Autonomous Drivingcited by 6
  26. 2024TempBEV: Improving Learned BEV Encoders with Combined Image and BEV Space Temporal Aggregationcited by 5
  27. 2024TEOcc: Radar-camera Multi-modal Occupancy Prediction via Temporal Enhancementcited by 5
  28. 2022BEV-LGKD: A Unified LiDAR-Guided Knowledge Distillation Framework for BEV 3D Object Detectioncited by 4
  29. 2023BEVStereo++: Accurate Depth Estimation in Multi-view 3D Object Detection via Dynamic Temporal Stereocited by 4
  30. 2026Towards Intrinsic-Aware Monocular 3D Object Detectioncited by 4
  31. 2023BEV-IO: Enhancing Bird's-Eye-View 3D Detection with Instance Occupancycited by 3
  32. 2023Instance-aware Multi-Camera 3D Object Detection with Structural Priors Mining and Self-Boosting Learningcited by 3
  33. 2024LSSInst: Improving Geometric Modeling in LSS-Based BEV Perception with Instance Representationcited by 2
  34. 2025CMF-IoU: Multi-Stage Cross-Modal Fusion 3D Object Detection with IoU Joint Predictioncited by 2
  35. 2024Learning Temporal Cues by Predicting Objects Move for Multi-camera 3D Object Detectioncited by 1
  36. 2024A Resource Efficient Fusion Network for Object Detection in Bird's-Eye View using Camera and Raw Radar Datacited by 1
  37. 2025REOcc: Camera-Radar Fusion with Radar Feature Enrichment for 3D Occupancy Predictioncited by 1
  38. 2023DFA3D: 3D Deformable Attention For 2D-to-3D Feature Liftingcited by 0

50 most cited of 53 All 53 (JSON)