mlsanityData to 7 Oct 2026

MaskLRF: Self-supervised Pretraining via Masked Autoencoding of Local Reference Frames for Rotation-invariant 3D Point Set Analysis

arXiv 2403.00206 ↗First posted 1 Mar 2024cs.CVcited by 4

Not on the map. Why?

Abstract (arXiv)

Following the successes in the fields of vision and language, self-supervised pretraining via masked autoencoding of 3D point set data, or Masked Point Modeling (MPM), has achieved state-of-the-art accuracy in various downstream tasks. However, current MPM methods lack a property essential for 3D point set analysis, namely, invariance against rotation of 3D objects/scenes. Existing MPM methods are thus not necessarily suitable for real-world applications where 3D point sets may have inconsistent orientations. This paper develops, for the first time, a rotation-invariant self-supervised pretraining framework for practical 3D point set analysis. The proposed algorithm, called MaskLRF, learns rotation-invariant and highly generalizable latent features via masked autoencoding of 3D points within Local Reference Frames (LRFs), which are not affected by rotation of 3D point sets. MaskLRF enhances the quality of latent features by integrating feature refinement using relative pose encoding and feature reconstruction using low-level but rich 3D geometry. The efficacy of MaskLRF is validated via extensive experiments on diverse downstream tasks including classification, segmentation, registration, and domain adaptation. I confirm that MaskLRF achieves new state-of-the-art accuracies in analyzing 3D point sets having inconsistent orientations. Code will be available at: https://github.com/takahikof/MaskLRF

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 63 of its 83 references that are in the map. First posted on arXiv 1 Mar 2024.

Builds on

63 of 83 references.

References · 63

  1. 2017Attention Is All You Needcited by 59,932
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  4. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  5. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  6. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  7. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  8. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  9. 2021SimMIM: A Simple Framework for Masked Image Modelingcited by 1,147
  10. 2021Point-BERT: Pre-training 3D Point Cloud Transformers with Masked Point Modelingcited by 647
  11. 2019Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Datacited by 532
  12. 2020PointContrast: Unsupervised Pre-training for 3D Point Cloud Understandingcited by 502
38 more
  1. 2021PointCLIP: Point Cloud Understanding by CLIPcited by 461
  2. 2022Masked Autoencoders for Point Cloud Self-supervised Learningcited by 456
  3. 20163DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructionscited by 333
  4. 2023OmniObject3D: Large-Vocabulary 3D Object Dataset for Realistic Perception, Reconstruction and Generationcited by 324
  5. 2021Vector Neurons: A General Framework for SO(3)-Equivariant Networkscited by 304
  6. 2020PREDATOR: Registration of 3D Point Clouds with Low Overlapcited by 287
  7. 2022Point-M2AE: Multi-scale Masked Autoencoders for Hierarchical Point Cloud Pre-trainingcited by 263
  8. 2021Self-Supervised Pretraining of 3D Features on any Point-Cloudcited by 217
  9. 2022Geometric Transformer for Fast and Robust Point Cloud Registrationcited by 200
  10. 2022CrossPoint: Self-Supervised Cross-Modal Contrastive Learning for 3D Point Cloud Understandingcited by 198
  11. 2020Unsupervised Point Cloud Pre-Training via Occlusion Completioncited by 186
  12. 2022Masked Discrimination for Self-Supervised Learning on Point Cloudscited by 155
  13. 2022Learning 3D Representations from 2D Pre-trained Models via Image-to-Point Masked Autoencoderscited by 143
  14. 2018PPF-FoldNet: Unsupervised Learning of Rotation Invariant 3D Local Descriptorscited by 140
  15. 2023PointGPT: Auto-regressively Generative Pre-training from Point Cloudscited by 115
  16. 2019PointDAN: A Multi-Scale 3D Domain Adaption Network for Point Cloud Representationcited by 114
  17. 2020Self-Supervised Learning for Domain Adaptation on Point-Cloudscited by 99
  18. 2019Rotation Invariant Convolutions for 3D Point Clouds Deep Learningcited by 80
  19. 2021PointCutMix: Regularization Strategy for Point Cloud Classificationcited by 69
  20. 2021You Only Hypothesize Once: Point Cloud Registration with Rotation-equivariant Descriptorscited by 68
  21. 2023Rotation-Invariant Transformer for Point Cloud Matchingcited by 68
  22. 2019Multi-Angle Point Cloud-VAE: Unsupervised Feature Learning for 3D Point Clouds from Multiple Angles by Joint Self-Reconstruction and Half-to-Half Predictioncited by 66
  23. 2020Self-Supervised Few-Shot Learning on Point Cloudscited by 66
  24. 2022Dilated Neighborhood Attention Transformercited by 64
  25. 2020A Rotation-Invariant Framework for Deep Point Cloud Analysiscited by 62
  26. 2022Occupancy-MAE: Self-supervised Pre-training Large-scale LiDAR Point Clouds with Masked Occupancy Autoencoderscited by 57
  27. 2022Unsupervised Point Cloud Representation Learning with Deep Neural Networks: A Surveycited by 54
  28. 2020Rotation-Invariant Local-to-Global Representation Learning for 3D Point Cloudcited by 38
  29. 2023GeoMAE: Masked Geometric Target Prediction for Self-supervised Point Cloud Pre-Trainingcited by 38
  30. 2021RandomRooms: Unsupervised Pre-training from Synthetic Shapes and Randomized Layouts for 3D Object Detectioncited by 37
  31. 2022Equivariant Point Cloud Analysis via Learning Orientations for Message Passingcited by 32
  32. 2021Geometry-Aware Self-Training for Unsupervised Domain Adaptationon Object Point Cloudscited by 31
  33. 20193D-Rotation-Equivariant Quaternion Neural Networkscited by 30
  34. 2019SRINet: Learning Strictly Rotation-Invariant Representations for Point Cloud Classification and Segmentationcited by 29
  35. 2022RIConv++: Effective Rotation Invariant Convolutions for 3D Point Clouds Deep Learningcited by 29
  36. 2020Learning to Orient Surfaces by Self-supervised Spherical CNNscited by 26
  37. 2021Domain Adaptation on Point Clouds via Geometry-Aware Implicitscited by 26
  38. 2023Bridging the Domain Gap: Self-Supervised 3D Scene Understanding with Foundation Modelscited by 26

50 most cited of 63 All 63 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 3; 2026: 1.

4 papers.

  1. 2025Token Adaptation via Side Graph Convolution for Efficient Fine-tuning of 3D Point Cloud Transformerscited by 2
  2. 2025HFBRI-MAE: Handcrafted Feature Based Rotation-Invariant Masked Autoencoder for 3D Point Cloud Analysiscited by 1
  3. 2025Beyond Random Masking: A Dual-Stream Approach for Rotation-Invariant Point Cloud Masked Autoencoderscited by 0
  4. 2026Data-Efficient Semantic Segmentation of 3D Point Clouds via Open-Vocabulary Image Segmentation-based Pseudo-Labelingcited by 0