mlsanityData to 7 Oct 2026

LaserMix for Semi-Supervised LiDAR Semantic Segmentation

arXiv 2207.00026 ↗First posted 30 Jun 2022cs.CV cs.LG cs.ROcited by 92

Not on the map. Why?

Abstract (arXiv)

Densely annotating LiDAR point clouds is costly, which restrains the scalability of fully-supervised learning methods. In this work, we study the underexplored semi-supervised learning (SSL) in LiDAR segmentation. Our core idea is to leverage the strong spatial cues of LiDAR point clouds to better exploit unlabeled data. We propose LaserMix to mix laser beams from different LiDAR scans, and then encourage the model to make consistent and confident predictions before and after mixing. Our framework has three appealing properties: 1) Generic: LaserMix is agnostic to LiDAR representations (e.g., range view and voxel), and hence our SSL framework can be universally applied. 2) Statistically grounded: We provide a detailed analysis to theoretically explain the applicability of the proposed framework. 3) Effective: Comprehensive experimental analysis on popular LiDAR segmentation datasets (nuScenes, SemanticKITTI, and ScribbleKITTI) demonstrates our effectiveness and superiority. Notably, we achieve competitive results over fully-supervised counterparts with 2x to 5x fewer labels and improve the supervised-only baseline significantly by 10.8% on average. We hope this concise yet high-performing framework could facilitate future research in semi-supervised LiDAR segmentation. Code is publicly available.

Key facts. Cited by 92 papers in the map (data through 7 Oct 2026). Builds on 47 of its 72 references that are in the map. First posted on arXiv 30 Jun 2022.

Builds on

47 of 72 references.

References · 47

  1. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  2. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  3. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  4. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  5. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
  6. 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,080
  7. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,623
  8. 2019MixMatch: A Holistic Approach to Semi-Supervised Learningcited by 1,390
  9. 2019KPConv: Flexible and Deformable Convolution for Point Cloudscited by 1,202
  10. 2016Temporal Ensembling for Semi-Supervised Learningcited by 1,054
  11. 2019RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Cloudscited by 584
  12. 2021Semi-Supervised Semantic Segmentation with Cross Pseudo Supervisioncited by 462
35 more
  1. 2020Searching Efficient 3D Architectures with Sparse Point-Voxel Convolutioncited by 422
  2. 2020Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentationcited by 388
  3. 2020Semi-Supervised Semantic Segmentation with Cross-Consistency Trainingcited by 375
  4. 2019ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoringcited by 349
  5. 2020PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentationcited by 274
  6. 2020SalsaNext: Fast, Uncertainty-aware Semantic Segmentation of LiDAR Point Clouds for Autonomous Drivingcited by 226
  7. 2020Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contextscited by 210
  8. 2020ClassMix: Segmentation-Based Data Augmentation for Semi-Supervised Learningcited by 209
  9. 2021ST++: Make Self-training Work Better for Semi-supervised Semantic Segmentationcited by 184
  10. 2020PseudoSeg: Designing Pseudo Labels for Semantic Segmentationcited by 180
  11. 2021RPVNet: A Deep and Efficient Range-Point-Voxel Fusion Network for LiDAR Point Cloud Segmentationcited by 173
  12. 2020SqueezeSegV3: Spatially-Adaptive Convolution for Efficient Point-Cloud Segmentationcited by 170
  13. 2021Panoptic nuScenes: A Large-Scale Benchmark for LiDAR Panoptic Segmentation and Trackingcited by 165
  14. 2021Offboard 3D Object Detection from Point Cloud Sequencescited by 144
  15. 2021Mix3D: Out-of-Context Data Augmentation for 3D Scenescited by 126
  16. 2020Guided Collaborative Training for Pixel-wise Semi-Supervised Learningcited by 120
  17. 2022Point-to-Voxel Knowledge Distillation for LiDAR Semantic Segmentationcited by 118
  18. 2021Perturbed and Strict Mean Teachers for Semi-supervised Semantic Segmentationcited by 113
  19. 2020AMVNet: Assertion-based Multi-View Fusion Network for LiDAR Semantic Segmentationcited by 87
  20. 2021Guided Point Contrastive Learning for Semi-supervised Point Cloud Semantic Segmentationcited by 87
  21. 2021A Simple Baseline for Semi-supervised Semantic Segmentation with Strong Data Augmentationcited by 79
  22. 2022Scribble-Supervised LiDAR Semantic Segmentationcited by 69
  23. 2021SQN: Weakly-Supervised Semantic Segmentation of Large-Scale 3D Point Cloudscited by 64
  24. 2021SSPC-Net: Semi-supervised Semantic 3D Point Cloud Segmentation Networkcited by 54
  25. 2021A Survey on Deep Domain Adaptation for LiDAR Perceptioncited by 46
  26. 2021FIDNet: LiDAR Point Cloud Semantic Segmentation with Fully Interpolation Decodingcited by 46
  27. 2019Liability, Ethics, and Culture-Aware Behavior Specification using Rulebookscited by 41
  28. 2021ConDA: Unsupervised Domain Adaptation for LiDAR Segmentation via Regularized Domain Concatenationcited by 41
  29. 2022DetMatch: Two Teachers are Better Than One for Joint 2D and 3D Semi-Supervised Object Detectioncited by 20
  30. 2018Semantic Segmentation of 3D LiDAR Data in Dynamic Scene Using Semi-supervised Learningcited by 19
  31. 2021Superpoint-guided Semi-supervised Semantic Segmentation of 3D Point Cloudscited by 19
  32. 2021PolarStream: Streaming Lidar Object Detection and Segmentation with Polar Pillarscited by 18
  33. 2022Box2Seg: Learning Semantics of 3D Point Clouds with Box-Level Supervisioncited by 11
  34. 2022Semi-supervised 3D Object Detection via Temporal Graph Neural Networkscited by 10
  35. 2022Semi-supervised 3D shape segmentation with multilevel consistency and part substitutioncited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 23; 2024: 34; 2025: 23; 2026: 11.

92 papers.

Sort:

By area:

  1. 2023CLIP2Scene: Towards Label-efficient 3D Scene Understanding by CLIPcited by 178
  2. 2023Robo3D: Towards Robust and Reliable 3D Perception against Corruptionscited by 128
  3. 2023Rethinking Range View Representation for LiDAR Segmentationcited by 126
  4. 2023Segment Any Point Cloud Sequences by Distilling Vision Foundation Modelscited by 123
  5. 20253D and 4D World Modeling: A Surveycited by 64
  6. 2023UniSeg: A Unified Multi-Modal LiDAR Segmentation Network and the OpenPCSeg Codebasecited by 57
  7. 2023Towards Label-free Scene Understanding by Vision Foundation Modelscited by 54
  8. 2024Multi-Modal Data-Efficient 3D Scene Understanding for Autonomous Drivingcited by 53
  9. 2023FRNet: Frustum-Range Networks for Scalable LiDAR Segmentationcited by 50
  10. 2023RoboDepth: Robust Out-of-Distribution Depth Estimation under Corruptionscited by 47
  11. 2024Benchmarking and Improving Bird's Eye View Perception Robustness in Autonomous Drivingcited by 39
  12. 2024A Survey on Occupancy Perception for Autonomous Driving: The Information Fusion Perspectivecited by 37
38 more
  1. 2023Less is More: Reducing Task and Model Complexity for 3D Point Cloud Semantic Segmentationcited by 33
  2. 20244D Contrastive Superflows are Dense 3D Representation Learnerscited by 32
  3. 2024Is Your HD Map Constructor Reliable under Sensor Corruptions?cited by 28
  4. 2024Is Your LiDAR Placement Optimized for 3D Scene Understanding?cited by 25
  5. 2023Advancements in Point Cloud Data Augmentation for Deep Learning: A Surveycited by 24
  6. 2024UniMix: Towards Domain Adaptive and Generalizable LiDAR Semantic Segmentation in Adverse Weathercited by 24
  7. 2024Multi-Space Alignments Towards Universal LiDAR Segmentationcited by 24
  8. 2023A Survey of Label-Efficient Deep Learning for 3D Point Cloudscited by 20
  9. 2023Learning to Adapt SAM for Segmenting Cross-domain Point Cloudscited by 20
  10. 2025LiMoE: Mixture of LiDAR Representation Learners from Automotive Scenescited by 19
  11. 2023Benchmarking the Robustness of LiDAR Semantic Segmentation Modelscited by 18
  12. 2024A Survey on Mixup Augmentations and Beyondcited by 18
  13. 2025Perspective-Invariant 3D Object Detectioncited by 16
  14. 2024Visual Foundation Models Boost Cross-Modal Unsupervised Domain Adaptation for 3D Semantic Segmentationcited by 15
  15. 2024RAPiD-Seg: Range-Aware Pointwise Distance Distribution Networks for 3D LiDAR Segmentationcited by 15
  16. 2024Rethinking Data Augmentation for Robust LiDAR Semantic Segmentation in Adverse Weathercited by 14
  17. 2023The RoboDepth Challenge: Methods and Advancements Towards Robust Depth Estimationcited by 13
  18. 2024TASeg: Temporal Aggregation Network for LiDAR Semantic Segmentationcited by 13
  19. 2024SFPNet: Sparse Focal Point Network for Semantic Segmentation on General LiDAR Point Cloudscited by 13
  20. 2025PointLoRA: Low-Rank Adaptation with Token Selection for Point Cloud Learningcited by 13
  21. 2023360^circ from a Single Camera: A Few-Shot Approach for LiDAR Segmentationcited by 12
  22. 2024An Empirical Study of Training State-of-the-Art LiDAR Segmentation Modelscited by 12
  23. 2023Annotator: A Generic Active Learning Baseline for LiDAR Semantic Segmentationcited by 11
  24. 2024Calib3D: Calibrating Model Preferences for Reliable 3D Scene Understandingcited by 11
  25. 2025LargeAD: Large-Scale Cross-Sensor Data Pretraining for Autonomous Drivingcited by 11
  26. 2025SPIRAL: Semantic-Aware Progressive LiDAR Scene Generation and Understandingcited by 10
  27. 2024Semi-supervised 3D Object Detection with PatchTeacher and PillarMixcited by 9
  28. 2024Building a Strong Pre-Training Baseline for Universal 3D Large-Scale Perceptioncited by 8
  29. 20253EED: Ground Everything Everywhere in 3Dcited by 8
  30. 2024Density-guided Translator Boosts Synthetic-to-Real Unsupervised Domain Adaptive Segmentation of 3D Point Cloudscited by 7
  31. 2024Hardness-Aware Scene Synthesis for Semi-Supervised 3D Object Detectioncited by 7
  32. 2025How Do Images Align and Complement LiDAR? Towards a Harmonized Multi-modal 3D Panoptic Segmentationcited by 7
  33. 2025Beyond One Shot, Beyond One Perspective: Cross-View and Long-Horizon Distillation for Better LiDAR Representationscited by 7
  34. 20232D Feature Distillation for Weakly- and Semi-Supervised 3D Semantic Segmentationcited by 6
  35. 2023Semi-Supervised Class-Agnostic Motion Prediction with Pseudo Label Regeneration and BEVMixcited by 6
  36. 2023Beyond the Label Itself: Latent Labels Enhance Semi-supervised Point Cloud Panoptic Segmentationcited by 6
  37. 2024Class-Imbalanced Semi-Supervised Learning for Large-Scale Point Cloud Semantic Segmentation via Decoupling Optimizationcited by 6
  38. 2023SPOT: Scalable 3D Pre-training via Occupancy Prediction for Learning Transferable 3D Representationscited by 5

50 most cited of 92 All 92 (JSON)