mlsanityData to 7 Oct 2026

Unsupervised Multi-Task Feature Learning on Point Clouds

arXiv 1910.08207 ↗First posted 18 Oct 2019cs.CV cs.LGcited by 89

Not on the map. Why?

Abstract (arXiv)

We introduce an unsupervised multi-task model to jointly learn point and shape features on point clouds. We define three unsupervised tasks including clustering, reconstruction, and self-supervised classification to train a multi-scale graph-based encoder. We evaluate our model on shape classification and segmentation benchmarks. The results suggest that it outperforms prior state-of-the-art unsupervised models: In the ModelNet40 classification task, it achieves an accuracy of 89.1% and in ShapeNet segmentation task, it achieves an mIoU of 68.2 and accuracy of 88.6%.

Key facts. Cited by 89 papers in the map (data through 7 Oct 2026). First posted on arXiv 18 Oct 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 21; 2022: 36; 2023: 18; 2024: 5; 2025: 6; 2026: 3.

89 papers.

Sort:

By area:

  1. 2021Self-Supervised Pretraining of 3D Features on any Point-Cloudcited by 217
  2. 2023A Survey on Self-supervised Learning: Algorithms, Applications, and Future Trendscited by 205
  3. 2022CrossPoint: Self-Supervised Cross-Modal Contrastive Learning for 3D Point Cloud Understandingcited by 198
  4. 2022Image-to-Lidar Self-Supervised Distillation for Autonomous Driving Datacited by 126
  5. 2022PartSLIP: Low-Shot Part Segmentation for 3D Point Clouds via Pretrained Image-Language Modelscited by 117
  6. 2023Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Trainingcited by 85
  7. 2022ALSO: Automotive Lidar Self-supervision by Occupancy estimationcited by 68
  8. 2022A Survey on Graph Neural Networks and Graph Transformers in Computer Vision: A Task-Oriented Perspectivecited by 56
  9. 2023Masked Scene Contrast: A Scalable Framework for Unsupervised 3D Representation Learningcited by 55
  10. 20214DContrast: Contrastive Learning with Dynamic Correspondences for 3D Scene Understandingcited by 54
  11. 2022Unsupervised Point Cloud Representation Learning with Deep Neural Networks: A Surveycited by 54
  12. 2021Self-Contrastive Learning with Hard Negative Sampling for Self-supervised Point Cloud Learningcited by 52
38 more
  1. 2022Implicit Autoencoder for Point-Cloud Self-Supervised Representation Learningcited by 46
  2. 2022Comprehensive Review of Deep Learning-Based 3D Point Cloud Completion Processing and Analysiscited by 44
  3. 2021RandomRooms: Unsupervised Pre-training from Synthetic Shapes and Randomized Layouts for 3D Object Detectioncited by 37
  4. 2021Learning from 2D: Contrastive Pixel-to-Point Knowledge Transfer for 3D Pretrainingcited by 32
  5. 2022Masked Autoencoders in 3D Point Cloud Representation Learningcited by 30
  6. 2022Weakly Supervised Semantic Segmentation for Large-Scale Point Cloudcited by 27
  7. 2023Bridging the Domain Gap: Self-Supervised 3D Scene Understanding with Foundation Modelscited by 26
  8. 2024GroupContrast: Semantic-aware Self-supervised Representation Learning for 3D Understandingcited by 26
  9. 2021Multimodal Semi-Supervised Learning for 3D Objectscited by 24
  10. 2022A Closer Look at Invariances in Self-supervised Pre-training for 3D Visioncited by 24
  11. 2022PedFormer: Pedestrian Behavior Prediction via Cross-Modal Attention Modulation and Gated Multitask Learningcited by 23
  12. 2023Self-supervised Learning for Pre-Training 3D Point Clouds: A Surveycited by 21
  13. 2022Cross-Domain Few-Shot Graph Classificationcited by 19
  14. 2023Learning to Estimate 6DoF Pose from Limited Data: A Few-Shot, Generalizable Approach using RGB Imagescited by 18
  15. 2023Lang3DSG: Language-based contrastive pre-training for 3D Scene Graph predictioncited by 18
  16. 2021Label-Efficient Point Cloud Semantic Segmentation: An Active Learning Approachcited by 17
  17. 2022LiDAL: Inter-frame Uncertainty Based Active Learning for 3D LiDAR Semantic Segmentationcited by 16
  18. 2022Masked Surfel Prediction for Self-Supervised Point Cloud Learningcited by 15
  19. 2022Learning Graph Augmentations to Learn Graph Representationscited by 14
  20. 2022Deep fiber clustering: Anatomically informed fiber clustering with self-supervised deep learning for fast and effective tractography parcellationcited by 13
  21. 2021HyperPocket: Generative Point Cloud Completioncited by 12
  22. 2022Data Efficient 3D Learner via Knowledge Transferred from 2D Modelcited by 10
  23. 2022MvDeCor: Multi-view Dense Correspondence Learning for Fine-grained 3D Segmentationcited by 10
  24. 2022CP-Net: Contour-Perturbed Reconstruction Network for Self-Supervised Point Cloud Learningcited by 9
  25. 2022Unsupervised Point Cloud Pre-Training via Contrasting and Clusteringcited by 9
  26. 2022Semi-supervised 3D shape segmentation with multilevel consistency and part substitutioncited by 8
  27. 2022Self-Supervised Learning with Multi-View Rendering for 3D Point Cloud Analysiscited by 8
  28. 2021Hierarchical View Predictor: Unsupervised 3D Global Feature Learning through Hierarchical Prediction among Unordered Viewscited by 7
  29. 2021Representing Shape Collections with Alignment-Aware Linear Modelscited by 7
  30. 2022PointVST: Self-Supervised Pre-training for 3D Point Clouds via View-Specific Point-to-Image Translationcited by 7
  31. 2021A Learnable Self-supervised Task for Unsupervised Domain Adaptation on Point Cloudscited by 6
  32. 2021PriFit: Learning to Fit Primitives Improves Few Shot Point Cloud Segmentationcited by 6
  33. 2022Data Augmentation-free Unsupervised Learning for 3D Point Cloud Understandingcited by 6
  34. 2022Randomized Quantization: A Generic Augmentation for Data Agnostic Self-supervised Learningcited by 6
  35. 2024DHGCN: Dynamic Hop Graph Convolution Network for Self-Supervised Point Cloud Learningcited by 6
  36. 2021Point Discriminative Learning for Data-efficient 3D Point Cloud Analysiscited by 5
  37. 2022PointCMC: Cross-Modal Multi-Scale Correspondences Learning for Point Cloud Understandingcited by 5
  38. 2022COLA: COarse LAbel pre-training for 3D semantic segmentation of sparse LiDAR datasetscited by 4

50 most cited of 89 All 89 (JSON)