mlsanityData to 7 Oct 2026

Mix-Teaching: A Simple, Unified and Effective Semi-Supervised Learning Framework for Monocular 3D Object Detection

arXiv 2207.04448 ↗First posted 10 Jul 2022cs.CVcited by 23

Not on the map. Why?

Abstract (arXiv)

Monocular 3D object detection is an essential perception task for autonomous driving. However, the high reliance on large-scale labeled data make it costly and time-consuming during model optimization. To reduce such over-reliance on human annotations, we propose Mix-Teaching, an effective semi-supervised learning framework applicable to employ both labeled and unlabeled images in training stage. Mix-Teaching first generates pseudo-labels for unlabeled images by self-training. The student model is then trained on the mixed images possessing much more intensive and precise labeling by merging instance-level image patches into empty backgrounds or labeled images. This is the first to break the image-level limitation and put high-quality pseudo labels from multi frames into one image for semi-supervised training. Besides, as a result of the misalignment between confidence score and localization quality, it's hard to discriminate high-quality pseudo-labels from noisy predictions using only confidence-based criterion. To that end, we further introduce an uncertainty-based filter to help select reliable pseudo boxes for the above mixing operation. To the best of our knowledge, this is the first unified SSL framework for monocular 3D object detection. Mix-Teaching consistently improves MonoFlex and GUPNet by significant margins under various labeling ratios on KITTI dataset. For example, our method achieves around +6.34% [email protected] improvement against the GUPNet baseline on validation set when using only 10% labeled data. Besides, by leveraging full training set and the additional 48K raw images of KITTI, it can further improve the MonoFlex by +4.65% improvement on [email protected] for car detection, reaching 18.54% [email protected], which ranks the 1st place among all monocular based methods on KITTI test leaderboard. The code and pretrained models will be released at https://github.com/yanglei18/Mix-Teaching.

Key facts. Cited by 23 papers in the map (data through 7 Oct 2026). Builds on 55 of its 80 references that are in the map. First posted on arXiv 10 Jul 2022.

Builds on

55 of 80 references.

References · 55

  1. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  2. 2019Scalability in Perception for Autonomous Driving: Waymo Open Datasetcited by 2,436
  3. 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,080
  4. 2018PointPillars: Fast Encoders for Object Detection from Point Cloudscited by 1,707
  5. 2019Self-training with Noisy Student improves ImageNet classificationcited by 1,304
  6. 2019Objects as Pointscited by 1,172
  7. 2021FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labelingcited by 537
  8. 2018Pseudo-LiDAR from Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Drivingcited by 455
  9. 2021Categorical Depth Distribution Network for Monocular 3D Object Detectioncited by 326
  10. 2021End-to-End Semi-Supervised Object Detection with Soft Teachercited by 315
  11. 2021Unbiased Teacher for Semi-Supervised Object Detectioncited by 305
  12. 2020A Simple Semi-Supervised Learning Framework for Object Detectioncited by 282
38 more
  1. 20163D Bounding Box Estimation Using Deep Learning and Geometrycited by 277
  2. 2021Is Pseudo-Lidar needed for Monocular 3D Object detection?cited by 248
  3. 2019Disentangling Monocular 3D Object Detectioncited by 215
  4. 2020SMOKE: Single-Stage Monocular 3D Object Detection via Keypoint Estimationcited by 202
  5. 2019Pseudo-LiDAR++: Accurate Depth for 3D Object Detection in Autonomous Drivingcited by 188
  6. 2021Objects are Different: Flexible Monocular 3D Object Detectioncited by 184
  7. 2019Learning Depth-Guided Convolutions for Monocular 3D Object Detectioncited by 169
  8. 2021Geometry Uncertainty Projection Network for Monocular 3D Object Detectioncited by 155
  9. 2020MonoPair: Monocular 3D Object Detection Using Pairwise Spatial Relationshipscited by 153
  10. 2020RTM3D: Real-time Monocular 3D Detection from Object Keypoints for Autonomous Drivingcited by 150
  11. 2021Delving into Localization Errors for Monocular 3D Object Detectioncited by 138
  12. 2019Accurate Monocular Object Detection via Color-Embedded 3D Reconstruction for Autonomous Drivingcited by 137
  13. 2021SE-SSD: Self-Ensembling Single-Stage Object Detector From Point Cloudcited by 119
  14. 2021Humble Teachers Teach Better Students for Semi-Supervised Object Detectioncited by 115
  15. 2021Instant-Teaching: An End-to-End Semi-Supervised Object Detection Frameworkcited by 112
  16. 2020Rethinking Pseudo-LiDAR Representationcited by 109
  17. 2022MonoDTR: Monocular 3D Object Detection with Depth-Aware Transformercited by 108
  18. 2019GS3D: An Efficient 3D Object Detection Framework for Autonomous Drivingcited by 106
  19. 2020ViP-DeepLab: Learning Visual Perception with Depth-aware Video Panoptic Segmentationcited by 95
  20. 2019SESS: Self-Ensembling Semi-Supervised 3D Object Detectioncited by 89
  21. 20203DIoUMatch: Leveraging IoU Prediction for Semi-Supervised 3D Object Detectioncited by 89
  22. 2021AutoShape: Real-Time Shape-Aware Monocular 3D Object Detectioncited by 87
  23. 2021Depth-conditioned Dynamic Message Propagation for Monocular 3D Object Detectioncited by 84
  24. 2021Geometry-based Distance Decomposition for Monocular 3D Object Detectioncited by 83
  25. 2021MonoRUn: Monocular 3D Object Detection by Reconstruction and Uncertainty Propagationcited by 82
  26. 2021Monocular 3D Object Detection: An Extrinsic Parameter Free Approachcited by 69
  27. 2022MonoDistill: Learning Spatial Features for Monocular 3D Object Detectioncited by 68
  28. 2019Deep Fitting Degree Scoring Network for Monocular 3D Object Detectioncited by 67
  29. 2021GrooMeD-NMS: Grouped Mathematically Differentiable NMS for Monocular 3D Object Detectioncited by 66
  30. 2021Ground-aware Monocular 3D Object Detection for Autonomous Drivingcited by 65
  31. 2020Temporal-Channel Transformer for 3D Lidar-Based Video Object Detection in Autonomous Drivingcited by 55
  32. 2021Progressive Coordinate Transforms for Monocular 3D Object Detectioncited by 52
  33. 2021Data-Uncertainty Guided Multi-Phase Learning for Semi-Supervised Object Detectioncited by 48
  34. 2021Rethinking Pseudo Labels for Semi-Supervised Object Detectioncited by 45
  35. 2020Interpolation-based semi-supervised learning for object detectioncited by 40
  36. 2021The Devil is in the Task: Exploiting Reciprocal Appearance-Localization Features for Monocular 3D Object Detectioncited by 34
  37. 2022MonoJSG: Joint Semantic and Geometric Cost Volume for Monocular 3D Object Detectioncited by 30
  38. 2021Lidar Point Cloud Guided Monocular 3D Object Detectioncited by 29

50 most cited of 55 All 55 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 9; 2024: 8; 2025: 3; 2026: 3.

23 papers.

Sort:

  1. 2023GraphAlign: Enhancing Accurate Feature Alignment by Graph matching for Multi-Modal 3D Object Detectioncited by 34
  2. 2023Dual Radar: A Multi-modal Dataset with Dual 4D Radar for Autonomous Drivingcited by 32
  3. 2023BEVHeight++: Toward Robust Visual Centric 3D Object Detectioncited by 29
  4. 2023Learning Cooperative Trajectory Representations for Motion Forecastingcited by 21
  5. 2024V2X-Radar: A Multi-modal Dataset with 4D Radar for Cooperative Perceptioncited by 21
  6. 2023MonoGAE: Roadside Monocular 3D Object Detection with Ground-Aware Embeddingscited by 20
  7. 2024VoxelNextFusion: A Simple, Unified and Effective Voxel Fusion Framework for Multi-Modal 3D Object Detectioncited by 14
  8. 2023NeurOCS: Neural NOCS Supervision for Monocular 3D Object Localizationcited by 13
  9. 2024CARLA Drone: Monocular 3D Object Detection from a Different Perspectivecited by 12
  10. 2024Decoupled Pseudo-labeling for Semi-Supervised Monocular 3D Object Detectioncited by 11
  11. 2023Semi-Supervised 2D Human Pose Estimation Driven by Position Inconsistency Pseudo Label Correction Modulecited by 10
  12. 2024MonoTAKD: Teaching Assistant Knowledge Distillation for Monocular 3D Object Detectioncited by 10
11 more
  1. 2023SSD-MonoDETR: Supervised Scale-aware Deformable Transformer for Monocular 3D Object Detectioncited by 9
  2. 2024Hardness-Aware Scene Synthesis for Semi-Supervised 3D Object Detectioncited by 7
  3. 2024SparseDet: A Simple and Effective Framework for Fully Sparse LiDAR-based 3D Object Detectioncited by 7
  4. 2024TiGDistill-BEV: Multi-view BEV 3D Object Detection via Target Inner-Geometry Learning Distillationcited by 4
  5. 2025MonoCT: Overcoming Monocular 3D Detection Domain Shift with Consistent Teacher Modelscited by 2
  6. 2025DGFusion: Dual-guided Fusion for Robust Multi-Modal 3D Object Detectioncited by 2
  7. 2025IDEAL-M3D: Instance Diversity-Enriched Active Learning for Monocular 3D Detectioncited by 2
  8. 2023Augment and Criticize: Exploring Informative Samples for Semi-Supervised Monocular 3D Object Detectioncited by 1
  9. 2026On the Assessment of Sensitivity of Autonomous Vehicle Perceptioncited by 1
  10. 2026MonoSAOD: Monocular 3D Object Detection with Sparsely Annotated Labelcited by 1
  11. 2026Selective Transfer Learning of Cross-Modality Distillation for Monocular 3D Object Detectioncited by 0