mlsanityData to 14 Oct 2026

Class-Level Confidence Based 3D Semi-Supervised Learning

arXiv 2210.10138 ↗First posted 18 Oct 2022cs.CV cs.AIcited by 14

Not on the map. Why?

Abstract (arXiv)

Recent state-of-the-art method FlexMatch firstly demonstrated that correctly estimating learning status is crucial for semi-supervised learning (SSL). However, the estimation method proposed by FlexMatch does not take into account imbalanced data, which is the common case for 3D semi-supervised learning. To address this problem, we practically demonstrate that unlabeled data class-level confidence can represent the learning status in the 3D imbalanced dataset. Based on this finding, we present a novel class-level confidence based 3D SSL method. Firstly, a dynamic thresholding strategy is proposed to utilize more unlabeled data, especially for low learning status classes. Then, a re-sampling strategy is designed to avoid biasing toward high learning status classes, which dynamically changes the sampling probability of each class. To show the effectiveness of our method in 3D SSL tasks, we conduct extensive experiments on 3D SSL classification and detection tasks. Our method significantly outperforms state-of-the-art counterparts for both 3D SSL classification and detection tasks in all datasets.

Key facts. Cited by 14 papers in the map (data through 14 Oct 2026). Builds on 37 of its 45 references that are in the map. First posted on arXiv 18 Oct 2022.

Builds on

37 of 45 references.

References · 37

  1. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  2. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  3. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,117
  4. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,223
  5. 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,085
  6. 2018PointPillars: Fast Encoders for Object Detection from Point Cloudscited by 1,715
  7. 2020Point Transformercited by 1,470
  8. 2019MixMatch: A Holistic Approach to Semi-Supervised Learningcited by 1,391
  9. 2020Big Self-Supervised Models are Strong Semi-Supervised Learnerscited by 1,380
  10. 2019KPConv: Flexible and Deformable Convolution for Point Cloudscited by 1,205
  11. 2018PointRCNN: 3D Object Proposal Generation and Detection from Point Cloudcited by 941
  12. 2017Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learningcited by 920
25 more
  1. 2020PCT: Point cloud transformercited by 732
  2. 2018PointConv: Deep Convolutional Networks on 3D Point Cloudscited by 657
  3. 2019Deep Hough Voting for 3D Object Detection in Point Cloudscited by 648
  4. 2017Frustum PointNets for 3D Object Detection from RGB-D Datacited by 640
  5. 2021FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labelingcited by 540
  6. 2019Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Datacited by 532
  7. 2019DeepGCNs: Can GCNs Go as Deep as CNNs?cited by 527
  8. 2019PIXOR: Real-time 3D Object Detection from Point Cloudscited by 358
  9. 2019ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoringcited by 349
  10. 2019S4L: Self-Supervised Semi-Supervised Learningcited by 335
  11. 2018SpiderCNN: Deep Learning on Point Sets with Parameterized Convolutional Filterscited by 261
  12. 2020Rethinking the Value of Labels for Improving Class-Imbalanced Learningcited by 216
  13. 2021CReST: A Class-Rebalancing Self-Training Framework for Imbalanced Semi-Supervised Learningcited by 175
  14. 2021Dash: Semi-Supervised Learning with Dynamic Thresholdingcited by 148
  15. 2015Deep Sliding Shapes for Amodal 3D Object Detection in RGB-D Imagescited by 111
  16. 2019GS3D: An Efficient 3D Object Detection Framework for Autonomous Drivingcited by 106
  17. 2019SESS: Self-Ensembling Semi-Supervised 3D Object Detectioncited by 89
  18. 20203DIoUMatch: Leveraging IoU Prediction for Semi-Supervised 3D Object Detectioncited by 89
  19. 2018Complex-YOLO: Real-time 3D Object Detection on Point Cloudscited by 57
  20. 2021PlayVirtual: Augmenting Cycle-Consistent Virtual Trajectories for Reinforcement Learningcited by 41
  21. 2019On the adequacy of untuned warmup for adaptive optimizationcited by 35
  22. 2020Class-Imbalanced Semi-Supervised Learningcited by 28
  23. 2021Multimodal Semi-Supervised Learning for 3D Objectscited by 24
  24. 2021Rethinking Re-Sampling in Imbalanced Semi-Supervised Learningcited by 16
  25. 2021On Data-Augmentation and Consistency-Based Semi-Supervised Learningcited by 13

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 3; 2024: 10; 2025: 0; 2026: 1.

14 papers.

Sort:

By area:

  1. 2023Bridging the Domain Gap: Self-Supervised 3D Scene Understanding with Foundation Modelscited by 27
  2. 2024MemoryMamba: Memory-Augmented State Space Model for Defect Recognitioncited by 24
  3. 2023CVRecon: Rethinking 3D Geometric Feature Learning For Neural Reconstructioncited by 16
  4. 2024Cross-Task Multi-Branch Vision Transformer for Facial Expression and Mask Wearing Classificationcited by 16
  5. 2024Incremental Joint Learning of Depth, Pose and Implicit Scene Representation on Monocular Camera in Large-scale Scenescited by 15
  6. 2024Adaptive speed planning for Unmanned Vehicle Based on Deep Reinforcement Learningcited by 15
  7. 2024TokenUnify: Scaling Up Autoregressive Pretraining for Neuron Segmentationcited by 7
  8. 2024Exploiting Diffusion Prior for Out-of-Distribution Detectioncited by 7
  9. 2024SAM-Guided Masked Token Prediction for 3D Scene Understandingcited by 3
  10. 2024Improving 3D Semi-supervised Learning by Effectively Utilizing All Unlabelled Datacited by 2
  11. 2023Unsupervised Gait Recognition with Selective Fusioncited by 1
  12. 2024DyConfidMatch: Dynamic Thresholding and Re-sampling for 3D Semi-supervised Learningcited by 1
2 more
  1. 2026An Adapter-free Fine-tuning Approach for Tuning 3D Foundation Modelscited by 1
  2. 2024TrajSSL: Trajectory-Enhanced Semi-Supervised 3D Object Detectioncited by 0