mlsanityData to 7 Oct 2026

Monocular 3D Object Detection with LiDAR Guided Semi Supervised Active Learning

arXiv 2307.08415 ↗First posted 17 Jul 2023cs.CVcited by 10

Not on the map. Why?

Abstract (arXiv)

We propose a novel semi-supervised active learning (SSAL) framework for monocular 3D object detection with LiDAR guidance (MonoLiG), which leverages all modalities of collected data during model development. We utilize LiDAR to guide the data selection and training of monocular 3D detectors without introducing any overhead in the inference phase. During training, we leverage the LiDAR teacher, monocular student cross-modal framework from semi-supervised learning to distill information from unlabeled data as pseudo-labels. To handle the differences in sensor characteristics, we propose a data noise-based weighting mechanism to reduce the effect of propagating noise from LiDAR modality to monocular. For selecting which samples to label to improve the model performance, we propose a sensor consistency-based selection score that is also coherent with the training objective. Extensive experimental results on KITTI and Waymo datasets verify the effectiveness of our proposed framework. In particular, our selection strategy consistently outperforms state-of-the-art active learning baselines, yielding up to 17% better saving rate in labeling costs. Our training strategy attains the top place in KITTI 3D and birds-eye-view (BEV) monocular object detection official benchmarks by improving the BEV Average Precision (AP) by 2.02.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Builds on 42 of its 58 references that are in the map. First posted on arXiv 17 Jul 2023.

Builds on

42 of 58 references.

References · 42

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  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. 2016Temporal Ensembling for Semi-Supervised Learningcited by 1,054
  6. 2019PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detectioncited by 890
  7. 2017Deep Bayesian Active Learning with Image Datacited by 812
  8. 2019Pseudo-Labeling and Confirmation Bias in Deep Semi-Supervised Learningcited by 507
  9. 2019Learning Loss for Active Learningcited by 335
  10. 2021Categorical Depth Distribution Network for Monocular 3D Object Detectioncited by 326
  11. 2021In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label Selection Framework for Semi-Supervised Learningcited by 301
  12. 2019Variational Adversarial Active Learningcited by 284
30 more
  1. 2019M3D-RPN: Monocular 3D Region Proposal Network for Object Detectioncited by 251
  2. 2021Is Pseudo-Lidar needed for Monocular 3D Object detection?cited by 248
  3. 2021Objects are Different: Flexible Monocular 3D Object Detectioncited by 184
  4. 2020Contextual Diversity for Active Learningcited by 139
  5. 2021Delving into Localization Errors for Monocular 3D Object Detectioncited by 138
  6. 2020A Review and Comparative Study on Probabilistic Object Detection in Autonomous Drivingcited by 115
  7. 2019Consistency-based Semi-supervised Active Learning: Towards Minimizing Labeling Costcited by 114
  8. 2022MonoDTR: Monocular 3D Object Detection with Depth-Aware Transformercited by 108
  9. 20203DIoUMatch: Leveraging IoU Prediction for Semi-Supervised 3D Object Detectioncited by 89
  10. 2021Depth-conditioned Dynamic Message Propagation for Monocular 3D Object Detectioncited by 84
  11. 2021MonoRUn: Monocular 3D Object Detection by Reconstruction and Uncertainty Propagationcited by 82
  12. 2021Active Learning for Deep Object Detection via Probabilistic Modelingcited by 79
  13. 2021Multiple instance active learning for object detectioncited by 77
  14. 2022MonoDistill: Learning Spatial Features for Monocular 3D Object Detectioncited by 68
  15. 2019Active Learning for Deep Detection Neural Networkscited by 62
  16. 2020Scalable Active Learning for Object Detectioncited by 58
  17. 2021Progressive Coordinate Transforms for Monocular 3D Object Detectioncited by 52
  18. 2018Evaluating Merging Strategies for Sampling-based Uncertainty Techniques in Object Detectioncited by 51
  19. 2021Semi-Supervised Active Learning with Temporal Output Discrepancycited by 50
  20. 2019Deep Active Learning for Efficient Training of a LiDAR 3D Object Detectorcited by 45
  21. 2019Rethinking deep active learning: Using unlabeled data at model trainingcited by 39
  22. 2022DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detectioncited by 39
  23. 2021Pseudo-labeling for Scalable 3D Object Detectioncited by 32
  24. 2021Consistency-based Active Learning for Object Detectioncited by 32
  25. 2021Lidar Point Cloud Guided Monocular 3D Object Detectioncited by 29
  26. 2021Not All Labels Are Equal: Rationalizing The Labeling Costs for Training Object Detectioncited by 27
  27. 2018Dropout-based Active Learning for Regressioncited by 24
  28. 2019Combining MixMatch and Active Learning for Better Accuracy with Fewer Labelscited by 17
  29. 2020Temporal Self-Ensembling Teacher for Semi-Supervised Object Detectioncited by 12
  30. 2020Semi-supervised Active Learning for Instance Segmentation via Scoring Predictionscited by 7

Built on by

Cited by year

Citing papers per year: 2024: 5; 2025: 5.

10 papers.

  1. 2024The Why, When, and How to Use Active Learning in Large-Data-Driven 3D Object Detection for Safe Autonomous Driving: An Empirical Explorationcited by 17
  2. 2024ActiveAnno3D -- An Active Learning Framework for Multi-Modal 3D Object Detectioncited by 17
  3. 2024Language-Driven Active Learning for Diverse Open-Set 3D Object Detectioncited by 15
  4. 2024A Unified Approach Towards Active Learning and Out-of-Distribution Detectioncited by 7
  5. 2025Joint Out-of-Distribution Filtering and Data Discovery Active Learningcited by 6
  6. 2024Reflective Teacher: Semi-Supervised Multimodal 3D Object Detection in Bird's-Eye-View via Uncertainty Measurecited by 2
  7. 2025IDEAL-M3D: Instance Diversity-Enriched Active Learning for Monocular 3D Detectioncited by 2
  8. 2025Inconsistency-based Active Learning for LiDAR Object Detectioncited by 1
  9. 2025ALScope: A Unified Toolkit for Deep Active Learningcited by 1
  10. 2025Evidential Deep Active Learning for Semi-Supervised Classificationcited by 0