mlsanityData to 14 Oct 2026

GroupContrast: Semantic-aware Self-supervised Representation Learning for 3D Understanding

arXiv 2403.09639 ↗First posted 14 Mar 2024cs.CVcited by 26

Not on the map. Why?

Abstract (arXiv)

Self-supervised 3D representation learning aims to learn effective representations from large-scale unlabeled point clouds. Most existing approaches adopt point discrimination as the pretext task, which assigns matched points in two distinct views as positive pairs and unmatched points as negative pairs. However, this approach often results in semantically identical points having dissimilar representations, leading to a high number of false negatives and introducing a "semantic conflict" problem. To address this issue, we propose GroupContrast, a novel approach that combines segment grouping and semantic-aware contrastive learning. Segment grouping partitions points into semantically meaningful regions, which enhances semantic coherence and provides semantic guidance for the subsequent contrastive representation learning. Semantic-aware contrastive learning augments the semantic information extracted from segment grouping and helps to alleviate the issue of "semantic conflict". We conducted extensive experiments on multiple 3D scene understanding tasks. The results demonstrate that GroupContrast learns semantically meaningful representations and achieves promising transfer learning performance.

Key facts. Cited by 26 papers in the map (data through 14 Oct 2026). Builds on 47 of its 58 references that are in the map. First posted on arXiv 14 Mar 2024.

Builds on

47 of 58 references.

References · 47

  1. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  2. 2023Segment Anythingcited by 8,234
  3. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  4. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  5. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  6. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  7. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  8. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,750
  9. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  10. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,117
  11. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,897
  12. 2020Exploring Simple Siamese Representation Learningcited by 2,866
35 more
  1. 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,857
  2. 2020Improved Baselines with Momentum Contrastive Learningcited by 2,287
  3. 2021An Empirical Study of Training Self-Supervised Vision Transformerscited by 1,616
  4. 2020Point Transformercited by 1,470
  5. 20194D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networkscited by 1,208
  6. 20173D Semantic Segmentation with Submanifold Sparse Convolutional Networkscited by 748
  7. 2019Deep Hough Voting for 3D Object Detection in Point Cloudscited by 648
  8. 2023Point Transformer V3: Simpler, Faster, Strongercited by 589
  9. 2020Dense Contrastive Learning for Self-Supervised Visual Pre-Trainingcited by 512
  10. 2020PointContrast: Unsupervised Pre-training for 3D Point Cloud Understandingcited by 504
  11. 2016Semantic Scene Completion from a Single Depth Imagecited by 472
  12. 2022Point Transformer V2: Grouped Vector Attention and Partition-based Poolingcited by 388
  13. 2019Deep Closest Point: Learning Representations for Point Cloud Registrationcited by 321
  14. 2020PointGroup: Dual-Set Point Grouping for 3D Instance Segmentationcited by 321
  15. 2014Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networkscited by 300
  16. 2020Propagate Yourself: Exploring Pixel-Level Consistency for Unsupervised Visual Representation Learningcited by 297
  17. 2022Language-Grounded Indoor 3D Semantic Segmentation in the Wildcited by 246
  18. 2021Self-Supervised Pretraining of 3D Features on any Point-Cloudcited by 217
  19. 2020Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contextscited by 210
  20. 2021Spatio-temporal Self-Supervised Representation Learning for 3D Point Cloudscited by 180
  21. 2023SAM3D: Segment Anything in 3D Scenescited by 157
  22. 2021Efficient Visual Pretraining with Contrastive Detectioncited by 150
  23. 2021Aligning Pretraining for Detection via Object-Level Contrastive Learningcited by 110
  24. 2019Unsupervised Multi-Task Feature Learning on Point Cloudscited by 89
  25. 2023Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Trainingcited by 86
  26. 2022Object discovery and representation networkscited by 80
  27. 2023UniPAD: A Universal Pre-training Paradigm for Autonomous Drivingcited by 73
  28. 2021Unsupervised Object-Level Representation Learning from Scene Imagescited by 72
  29. 2020Self-EMD: Self-Supervised Object Detection without ImageNetcited by 70
  30. 2022Self-Supervised Visual Representation Learning with Semantic Groupingcited by 70
  31. 2020Info3D: Representation Learning on 3D Objects using Mutual Information Maximization and Contrastive Learningcited by 68
  32. 2023Masked Scene Contrast: A Scalable Framework for Unsupervised 3D Representation Learningcited by 56
  33. 2023PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigmcited by 56
  34. 2023Understanding Imbalanced Semantic Segmentation Through Neural Collapsecited by 46
  35. 2023Clustering based Point Cloud Representation Learning for 3D Analysiscited by 23

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 5; 2025: 13; 2026: 8.

26 papers.

Sort:

By area:

  1. 2025Sonata: Self-Supervised Learning of Reliable Point Representationscited by 110
  2. 2025Concerto: Joint 2D-3D Self-Supervised Learning Emerges Spatial Representationscited by 45
  3. 2024OA-CNNs: Omni-Adaptive Sparse CNNs for 3D Semantic Segmentationcited by 44
  4. 2026FlashVID: Efficient Video Large Language Models via Training-free Tree-based Spatiotemporal Token Mergingcited by 18
  5. 2025Less Is More, but Where? Dynamic Token Compression via LLM-Guided Keyframe Priorcited by 15
  6. 2024LCM: Locally Constrained Compact Point Cloud Model for Masked Point Modelingcited by 12
  7. 2025Masked Point-Entity Contrast for Open-Vocabulary 3D Scene Understandingcited by 10
  8. 2025UniPre3D: Unified Pre-training of 3D Point Cloud Models with Cross-Modal Gaussian Splattingcited by 6
  9. 2024Fine-grained Image-to-LiDAR Contrastive Distillation with Visual Foundation Modelscited by 5
  10. 2025Structure-aware Semantic Discrepancy and Consistency for 3D Medical Image Self-supervised Learningcited by 5
  11. 2026Dynamic-dLLM: Dynamic Cache-Budget and Adaptive Parallel Decoding for Training-Free Acceleration of Diffusion LLMcited by 5
  12. 2025Masked Scene Modeling: Narrowing the Gap Between Supervised and Self-Supervised Learning in 3D Scene Understandingcited by 4
14 more
  1. 2025Forging Spatial Intelligence: A Roadmap of Multi-Modal Data Pre-Training for Autonomous Systemscited by 4
  2. 2024Exploring Scene Affinity for Semi-Supervised LiDAR Semantic Segmentationcited by 2
  3. 2025Token Adaptation via Side Graph Convolution for Efficient Fine-tuning of 3D Point Cloud Transformerscited by 2
  4. 2025GaussianCross: Cross-modal Self-supervised 3D Representation Learning via Gaussian Splattingcited by 2
  5. 2024Semantic-Free Procedural 3D Shapes Are Surprisingly Good Teacherscited by 1
  6. 2026PointCSP: Cross-Sample Semantic Propagation and Stability Preservation in Self-Supervised Point Cloud Learningcited by 1
  7. 2025CoStoDet-DDPM: Collaborative Training of Stochastic and Deterministic Models Improves Surgical Workflow Anticipation and Recognitioncited by 0
  8. 2025Self-Supervised Moving Object Segmentation of Sparse and Noisy Radar Point Cloudscited by 0
  9. 20253D sans 3D Scans: Scalable Pre-training from Video-Generated Point Cloudscited by 0
  10. 2026P3T: Prototypical Point-level Prompt Tuning with Enhanced Generalization for 3D Vision-Language Modelscited by 0
  11. 2026Beyond Binary Contrast: Modeling Continuous Skeleton Action Spaces with Transitional Anchorscited by 0
  12. 2026Geometry-Guided Self-Supervision for Ultra-Fine-Grained Recognition with Limited Datacited by 0
  13. 2026DC-Leap: Training-Free Acceleration of dLLMs via Draft-Guided Contiguous Leaping Decodingcited by 0
  14. 2026PAGER: Partial-to-global Alignment via Geometric and Relational Distillationcited by 0