mlsanityData to 7 Oct 2026

Invariant Training 2D-3D Joint Hard Samples for Few-Shot Point Cloud Recognition

arXiv 2308.09694 ↗First posted 18 Aug 2023cs.CVcited by 13

Not on the map. Why?

Abstract (arXiv)

We tackle the data scarcity challenge in few-shot point cloud recognition of 3D objects by using a joint prediction from a conventional 3D model and a well-trained 2D model. Surprisingly, such an ensemble, though seems trivial, has hardly been shown effective in recent 2D-3D models. We find out the crux is the less effective training for the ''joint hard samples'', which have high confidence prediction on different wrong labels, implying that the 2D and 3D models do not collaborate well. To this end, our proposed invariant training strategy, called InvJoint, does not only emphasize the training more on the hard samples, but also seeks the invariance between the conflicting 2D and 3D ambiguous predictions. InvJoint can learn more collaborative 2D and 3D representations for better ensemble. Extensive experiments on 3D shape classification with widely adopted ModelNet10/40, ScanObjectNN and Toys4K, and shape retrieval with ShapeNet-Core validate the superiority of our InvJoint.

Key facts. Cited by 13 papers in the map (data through 7 Oct 2026). Builds on 56 of its 81 references that are in the map. First posted on arXiv 18 Aug 2023.

Builds on

56 of 81 references.

References · 56

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  4. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  5. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  6. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  7. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  8. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  9. 2022BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generationcited by 4,310
  10. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  11. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  12. 2020Supervised Contrastive Learningcited by 2,803
38 more
  1. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  2. 2019ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Taskscited by 2,219
  3. 2019Invariant Risk Minimizationcited by 1,832
  4. 2020Point Transformercited by 1,462
  5. 2021CLIP-Adapter: Better Vision-Language Models with Feature Adapterscited by 1,021
  6. 2017In Defense of the Triplet Loss for Person Re-Identificationcited by 784
  7. 2018PointCNN: Convolution On X-Transformed Pointscited by 764
  8. 2020Out-of-Distribution Generalization via Risk Extrapolation (REx)cited by 737
  9. 2020PCT: Point cloud transformercited by 730
  10. 2020DivideMix: Learning with Noisy Labels as Semi-supervised Learningcited by 691
  11. 2021Point-BERT: Pre-training 3D Point Cloud Transformers with Masked Point Modelingcited by 647
  12. 2016Training Region-based Object Detectors with Online Hard Example Miningcited by 577
  13. 2015Multi-view Convolutional Neural Networks for 3D Shape Recognitioncited by 564
  14. 2019Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Datacited by 532
  15. 2020PointContrast: Unsupervised Pre-training for 3D Point Cloud Understandingcited by 502
  16. 2021PointCLIP: Point Cloud Understanding by CLIPcited by 461
  17. 2022Masked Autoencoders for Point Cloud Self-supervised Learningcited by 456
  18. 2022PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategiescited by 432
  19. 2020Voxel R-CNN: Towards High Performance Voxel-based 3D Object Detectioncited by 409
  20. 2022Rethinking Network Design and Local Geometry in Point Cloud: A Simple Residual MLP Frameworkcited by 376
  21. 2019SynSin: End-to-end View Synthesis from a Single Imagecited by 364
  22. 2023OmniObject3D: Large-Vocabulary 3D Object Dataset for Realistic Perception, Reconstruction and Generationcited by 324
  23. 2018SpiderCNN: Deep Learning on Point Sets with Parameterized Convolutional Filterscited by 261
  24. 2023MVImgNet: A Large-scale Dataset of Multi-view Imagescited by 219
  25. 2022CrossPoint: Self-Supervised Cross-Modal Contrastive Learning for 3D Point Cloud Understandingcited by 198
  26. 2020Unsupervised Point Cloud Pre-Training via Occlusion Completioncited by 186
  27. 2021Revisiting Point Cloud Shape Classification with a Simple and Effective Baselinecited by 145
  28. 2020MVTN: Multi-View Transformation Network for 3D Shape Recognitioncited by 140
  29. 2021Using Shape to Categorize: Low-Shot Learning with an Explicit Shape Biascited by 115
  30. 2019DensePoint: Learning Densely Contextual Representation for Efficient Point Cloud Processingcited by 104
  31. 2020Convolutional Networks with Dense Connectivitycited by 98
  32. 2022Benchmarking Robustness of 3D Point Cloud Recognition Against Common Corruptionscited by 78
  33. 2022P2P: Tuning Pre-trained Image Models for Point Cloud Analysis with Point-to-Pixel Promptingcited by 74
  34. 2019A-CNN: Annularly Convolutional Neural Networks on Point Cloudscited by 66
  35. 2020Self-Supervised Few-Shot Learning on Point Cloudscited by 66
  36. 2021Adaptive Graph Convolution for Point Cloud Analysiscited by 60
  37. 2021Image2Point: 3D Point-Cloud Understanding with 2D Image Pretrained Modelscited by 41
  38. 2018PVNet: A Joint Convolutional Network of Point Cloud and Multi-View for 3D Shape Recognitioncited by 36

50 most cited of 56 All 56 (JSON)

Built on by

Cited by year

Citing papers per year: 2023: 2; 2024: 10; 2025: 1.

13 papers.

Sort:

By area:

  1. 2024View-Consistent 3D Editing with Gaussian Splattingcited by 67
  2. 2024Consistent3D: Towards Consistent High-Fidelity Text-to-3D Generation with Deterministic Sampling Priorcited by 60
  3. 2024Gamba: Marry Gaussian Splatting with Mamba for single view 3D reconstructioncited by 56
  4. 2023Generalized Logit Adjustment: Calibrating Fine-tuned Models by Removing Label Bias in Foundation Modelscited by 32
  5. 2024HumanEdit: A High-Quality Human-Rewarded Dataset for Instruction-based Image Editingcited by 29
  6. 2024Enhancing Zero-Shot Vision Models by Label-Free Prompt Distribution Learning and Bias Correctingcited by 26
  7. 2024MVGamba: Unify 3D Content Generation as State Space Sequence Modelingcited by 23
  8. 2024It's Never Too Late: Fusing Acoustic Information into Large Language Models for Automatic Speech Recognitioncited by 22
  9. 2024Diffusion Time-step Curriculum for One Image to 3D Generationcited by 20
  10. 2023Integrating View Conditions for Image Synthesiscited by 16
  11. 2025Seeing World Dynamics in a Nutshellcited by 7
  12. 2024Formula-Supervised Visual-Geometric Pre-trainingcited by 3
1 more
  1. 2024Efficient 3D Perception on Multi-Sweep Point Cloud with Gumbel Spatial Pruningcited by 1