mlsanityData to 7 Oct 2026

CoSMix: Compositional Semantic Mix for Domain Adaptation in 3D LiDAR Segmentation

arXiv 2207.09778 ↗First posted 20 Jul 2022cs.CV cs.AI cs.LGcited by 54

Not on the map. Why?

Abstract (arXiv)

3D LiDAR semantic segmentation is fundamental for autonomous driving. Several Unsupervised Domain Adaptation (UDA) methods for point cloud data have been recently proposed to improve model generalization for different sensors and environments. Researchers working on UDA problems in the image domain have shown that sample mixing can mitigate domain shift. We propose a new approach of sample mixing for point cloud UDA, namely Compositional Semantic Mix (CoSMix), the first UDA approach for point cloud segmentation based on sample mixing. CoSMix consists of a two-branch symmetric network that can process labelled synthetic data (source) and real-world unlabelled point clouds (target) concurrently. Each branch operates on one domain by mixing selected pieces of data from the other one, and by using the semantic information derived from source labels and target pseudo-labels. We evaluate CoSMix on two large-scale datasets, showing that it outperforms state-of-the-art methods by a large margin. Our code is available at https://github.com/saltoricristiano/cosmix-uda.

Key facts. Cited by 54 papers in the map (data through 7 Oct 2026). Builds on 37 of its 46 references that are in the map. First posted on arXiv 20 Jul 2022.

Builds on

37 of 46 references.

References · 37

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  2. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  3. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  4. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  5. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  6. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
  7. 2017CARLA: An Open Urban Driving Simulatorcited by 2,431
  8. 2017VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detectioncited by 1,323
  9. 2017Adversarial Discriminative Domain Adaptationcited by 1,309
  10. 20194D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networkscited by 1,201
  11. 20173D Semantic Segmentation with Submanifold Sparse Convolutional Networkscited by 746
  12. 2018ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentationcited by 623
25 more
  1. 2019RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Cloudscited by 584
  2. 2020FDA: Fourier Domain Adaptation for Semantic Segmentationcited by 540
  3. 2019Confidence Regularized Self-Trainingcited by 418
  4. 2020Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentationcited by 388
  5. 2020PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentationcited by 274
  6. 2019Adversarial Domain Adaptation with Domain Mixupcited by 250
  7. 2017SqueezeSeg: Convolutional Neural Nets with Recurrent CRF for Real-Time Road-Object Segmentation from 3D LiDAR Point Cloudcited by 231
  8. 2018SqueezeSegV2: Improved Model Structure and Unsupervised Domain Adaptation for Road-Object Segmentation from a LiDAR Point Cloudcited by 228
  9. 2020DACS: Domain Adaptation via Cross-domain Mixed Samplingcited by 226
  10. 2017Submanifold Sparse Convolutional Networkscited by 223
  11. 2020A survey of loss functions for semantic segmentationcited by 216
  12. 2019xMUDA: Cross-Modal Unsupervised Domain Adaptation for 3D Semantic Segmentationcited by 129
  13. 2017Domain Adaptation for Visual Applications: A Comprehensive Surveycited by 127
  14. 2021Mix3D: Out-of-Context Data Augmentation for 3D Scenescited by 126
  15. 2020SemanticPOSS: A Point Cloud Dataset with Large Quantity of Dynamic Instancescited by 123
  16. 2020Complete & Label: A Domain Adaptation Approach to Semantic Segmentation of LiDAR Point Cloudscited by 116
  17. 2020PointMixup: Augmentation for Point Cloudscited by 104
  18. 2020Dual Mixup Regularized Learning for Adversarial Domain Adaptationcited by 99
  19. 2021Domain Adaptive Semantic Segmentation with Self-Supervised Depth Estimationcited by 80
  20. 2021PointCutMix: Regularization Strategy for Point Cloud Classificationcited by 69
  21. 20203D-MiniNet: Learning a 2D Representation from Point Clouds for Fast and Efficient 3D LIDAR Semantic Segmentationcited by 58
  22. 2020Towards Recognizing Unseen Categories in Unseen Domainscited by 56
  23. 2020ePointDA: An End-to-End Simulation-to-Real Domain Adaptation Framework for LiDAR Point Cloud Segmentationcited by 56
  24. 2021DSP: Dual Soft-Paste for Unsupervised Domain Adaptive Semantic Segmentationcited by 45
  25. 2021Geometry-Aware Self-Training for Unsupervised Domain Adaptationon Object Point Cloudscited by 31

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 25; 2024: 11; 2025: 9; 2026: 6.

54 papers.

Sort:

By area:

  1. 2022PolarMix: A General Data Augmentation Technique for LiDAR Point Cloudscited by 74
  2. 20233D Semantic Segmentation in the Wild: Learning Generalized Models for Adverse-Condition Point Cloudscited by 46
  3. 2023Three Pillars improving Vision Foundation Model Distillation for Lidarcited by 43
  4. 2022Domain generalization of 3D semantic segmentation in autonomous drivingcited by 25
  5. 2024UniMix: Towards Domain Adaptive and Generalizable LiDAR Semantic Segmentation in Adverse Weathercited by 24
  6. 2024Multi-Space Alignments Towards Universal LiDAR Segmentationcited by 24
  7. 2023Automated forest inventory: analysis of high-density airborne LiDAR point clouds with 3D deep learningcited by 23
  8. 2023Novel Class Discovery for 3D Point Cloud Semantic Segmentationcited by 21
  9. 2023SALUDA: Surface-based Automotive Lidar Unsupervised Domain Adaptationcited by 21
  10. 2023A Survey of Label-Efficient Deep Learning for 3D Point Cloudscited by 20
  11. 2023Compositional Semantic Mix for Domain Adaptation in Point Cloud Segmentationcited by 20
  12. 2023Learning to Adapt SAM for Segmenting Cross-domain Point Cloudscited by 20
38 more
  1. 2023S2R-ViT for Multi-Agent Cooperative Perception: Bridging the Gap from Simulation to Realitycited by 18
  2. 2023MoPA: Multi-Modal Prior Aided Domain Adaptation for 3D Semantic Segmentationcited by 17
  3. 2023Instant Domain Augmentation for LiDAR Semantic Segmentationcited by 16
  4. 2024Visual Foundation Models Boost Cross-Modal Unsupervised Domain Adaptation for 3D Semantic Segmentationcited by 15
  5. 2023Exploiting the Complementarity of 2D and 3D Networks to Address Domain-Shift in 3D Semantic Segmentationcited by 14
  6. 2024Backpropagation-free Network for 3D Test-time Adaptationcited by 14
  7. 2024Rethinking Data Augmentation for Robust LiDAR Semantic Segmentation in Adverse Weathercited by 14
  8. 2023Walking Your LiDOG: A Journey Through Multiple Domains for LiDAR Semantic Segmentationcited by 13
  9. 2023360^circ from a Single Camera: A Few-Shot Approach for LiDAR Segmentationcited by 12
  10. 2023Annotator: A Generic Active Learning Baseline for LiDAR Semantic Segmentationcited by 11
  11. 2023GeoAdapt: Self-Supervised Test-Time Adaptation in LiDAR Place Recognition Using Geometric Priorscited by 8
  12. 2023Rethinking LiDAR Domain Generalization: Single Source as Multiple Density Domainscited by 7
  13. 2024Density-guided Translator Boosts Synthetic-to-Real Unsupervised Domain Adaptive Segmentation of 3D Point Cloudscited by 7
  14. 2023You Never Get a Second Chance To Make a Good First Impression: Seeding Active Learning for 3D Semantic Segmentationcited by 6
  15. 2024DiffuBox: Refining 3D Object Detection with Point Diffusioncited by 6
  16. 2024Train Till You Drop: Towards Stable and Robust Source-free Unsupervised 3D Domain Adaptationcited by 6
  17. 2023T-UDA: Temporal Unsupervised Domain Adaptation in Sequential Point Cloudscited by 5
  18. 2023ParisLuco3D: A high-quality target dataset for domain generalization of LiDAR perceptioncited by 5
  19. 2024TTT-KD: Test-Time Training for 3D Semantic Segmentation through Knowledge Distillation from Foundation Modelscited by 5
  20. 2023Cross-modal & Cross-domain Learning for Unsupervised LiDAR Semantic Segmentationcited by 4
  21. 2023Novel class discovery meets foundation models for 3D semantic segmentationcited by 4
  22. 2024Text3DAug -- Prompted Instance Augmentation for LiDAR Perceptioncited by 4
  23. 2024Fusion-then-Distillation: Toward Cross-modal Positive Distillation for Domain Adaptive 3D Semantic Segmentationcited by 4
  24. 2023MixCycle: Mixup Assisted Semi-Supervised 3D Single Object Tracking with Cycle Consistencycited by 3
  25. 2023Model2Scene: Learning 3D Scene Representation via Contrastive Language-CAD Models Pre-trainingcited by 3
  26. 2025SAM-guided Pseudo Label Enhancement for Multi-modal 3D Semantic Segmentationcited by 3
  27. 2022Fake it, Mix it, Segment it: Bridging the Domain Gap Between Lidar Sensorscited by 2
  28. 2025APCoTTA: Continual Test-Time Adaptation for Semantic Segmentation of Airborne LiDAR Point Cloudscited by 2
  29. 2025Heterogeneous-Modal Unsupervised Domain Adaptation via Latent Space Bridgingcited by 2
  30. 2025SG-LDM: Semantic-Guided LiDAR Generation via Latent-Aligned Diffusioncited by 2
  31. 2025Improving Multimodal Distillation for 3D Semantic Segmentation under Domain Shiftcited by 2
  32. 2023Paved2Paradise: Cost-Effective and Scalable LiDAR Simulation by Factoring the Real Worldcited by 1
  33. 2025WHU-STree: A Multi-modal Benchmark Dataset for Street Tree Inventorycited by 1
  34. 20253D Weakly Supervised Semantic Segmentation via Class-Aware and Geometry-Guided Pseudo-Label Refinementcited by 1
  35. 2025DPGLA: Bridging the Gap between Synthetic and Real Data for Unsupervised Domain Adaptation in 3D LiDAR Semantic Segmentationcited by 1
  36. 2026Source-Free Domain Adaptation for Geospatial Point Cloud Semantic Segmentationcited by 1
  37. 2026TOLiD: Bridging the Architecture Gap in Vision Foundation Model to LiDAR Pretraining via Token Lifting for Distillationcited by 1
  38. 2025RAFT -- A Domain Adaptation Framework for RGB & LiDAR Semantic Segmentationcited by 0

50 most cited of 54 All 54 (JSON)