mlsanityData to 7 Oct 2026

PointGroup: Dual-Set Point Grouping for 3D Instance Segmentation

arXiv 2004.01658 ↗First posted 3 Apr 2020cs.CVcited by 318

Not on the map. Why?

Abstract (arXiv)

Instance segmentation is an important task for scene understanding. Compared to the fully-developed 2D, 3D instance segmentation for point clouds have much room to improve. In this paper, we present PointGroup, a new end-to-end bottom-up architecture, specifically focused on better grouping the points by exploring the void space between objects. We design a two-branch network to extract point features and predict semantic labels and offsets, for shifting each point towards its respective instance centroid. A clustering component is followed to utilize both the original and offset-shifted point coordinate sets, taking advantage of their complementary strength. Further, we formulate the ScoreNet to evaluate the candidate instances, followed by the Non-Maximum Suppression (NMS) to remove duplicates. We conduct extensive experiments on two challenging datasets, ScanNet v2 and S3DIS, on which our method achieves the highest performance, 63.6% and 64.0%, compared to 54.9% and 54.4% achieved by former best solutions in terms of mAP with IoU threshold 0.5.

Key facts. Cited by 318 papers in the map (data through 7 Oct 2026). First posted on arXiv 3 Apr 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 40; 2022: 46; 2023: 79; 2024: 60; 2025: 62; 2026: 31.

318 papers.

Sort:

By area:

  1. 2023ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenescited by 747
  2. 2023Point Transformer V3: Simpler, Faster, Strongercited by 579
  3. 2021KITTI-360: A Novel Dataset and Benchmarks for Urban Scene Understanding in 2D and 3Dcited by 570
  4. 2022Mask3D: Mask Transformer for 3D Semantic Instance Segmentationcited by 299
  5. 2023OpenMask3D: Open-Vocabulary 3D Instance Segmentationcited by 288
  6. 2022Language-Grounded Indoor 3D Semantic Segmentation in the Wildcited by 243
  7. 20233D-VisTA: Pre-trained Transformer for 3D Vision and Text Alignmentcited by 201
  8. 2022SoftGroup for 3D Instance Segmentation on Point Cloudscited by 197
  9. 2022PLA: Language-Driven Open-Vocabulary 3D Scene Understandingcited by 176
  10. 2024Aligning Cyber Space with Physical World: A Comprehensive Survey on Embodied AIcited by 175
  11. 2023Chat-Scene: Bridging 3D Scene and Large Language Models with Object Identifierscited by 170
  12. 2024Scene-LLM: Extending Language Model for 3D Visual Understanding and Reasoningcited by 170
38 more
  1. 2023Multi3DRefer: Grounding Text Description to Multiple 3D Objectscited by 142
  2. 2021InstanceRefer: Cooperative Holistic Understanding for Visual Grounding on Point Clouds through Instance Multi-level Contextual Referringcited by 140
  3. 2022Multi-View Transformer for 3D Visual Groundingcited by 138
  4. 2022GAPartNet: Cross-Category Domain-Generalizable Object Perception and Manipulation via Generalizable and Actionable Partscited by 135
  5. 2024SceneVerse: Scaling 3D Vision-Language Learning for Grounded Scene Understandingcited by 130
  6. 2021Hierarchical Aggregation for 3D Instance Segmentationcited by 126
  7. 2023Chat-3D: Data-efficiently Tuning Large Language Model for Universal Dialogue of 3D Scenescited by 125
  8. 2021SAT: 2D Semantics Assisted Training for 3D Visual Groundingcited by 124
  9. 2023Transformer-Based Visual Segmentation: A Surveycited by 124
  10. 2023Open3DIS: Open-Vocabulary 3D Instance Segmentation with 2D Mask Guidancecited by 123
  11. 2022Language Conditioned Spatial Relation Reasoning for 3D Object Groundingcited by 121
  12. 2022PartSLIP: Low-Shot Part Segmentation for 3D Point Clouds via Pretrained Image-Language Modelscited by 117
  13. 2022Superpoint Transformer for 3D Scene Instance Segmentationcited by 115
  14. 2024Grounded 3D-LLM with Referent Tokenscited by 111
  15. 2025Sonata: Self-Supervised Learning of Reliable Point Representationscited by 108
  16. 2025GPT4Scene: Understand 3D Scenes from Videos with Vision-Language Modelscited by 106
  17. 2021Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networkscited by 98
  18. 2022EDA: Explicit Text-Decoupling and Dense Alignment for 3D Visual Groundingcited by 98
  19. 2023OneFormer3D: One Transformer for Unified Point Cloud Segmentationcited by 95
  20. 2023RegionPLC: Regional Point-Language Contrastive Learning for Open-World 3D Scene Understandingcited by 93
  21. 2025PARTFIELD: Learning 3D Feature Fields for Part Segmentation and Beyondcited by 91
  22. 2021One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentationcited by 89
  23. 2021Guided Point Contrastive Learning for Semi-supervised Point Cloud Semantic Segmentationcited by 87
  24. 2023Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Trainingcited by 85
  25. 2022Fully Sparse 3D Object Detectioncited by 84
  26. 2023End-to-End 3D Dense Captioning with Vote2Cap-DETRcited by 77
  27. 2021EfficientLPS: Efficient LiDAR Panoptic Segmentationcited by 76
  28. 2021Panoptic 3D Scene Reconstruction From a Single RGB Imagecited by 70
  29. 2023ViewRefer: Grasp the Multi-view Knowledge for 3D Visual Grounding with GPT and Prototype Guidancecited by 70
  30. 2024Unifying 3D Vision-Language Understanding via Promptable Queriescited by 70
  31. 2022UniT3D: A Unified Transformer for 3D Dense Captioning and Visual Groundingcited by 69
  32. 20214D Panoptic LiDAR Segmentationcited by 64
  33. 2023Visual Programming for Zero-shot Open-Vocabulary 3D Visual Groundingcited by 64
  34. 2023Segment3D: Learning Fine-Grained Class-Agnostic 3D Segmentation without Manual Labelscited by 64
  35. 2021Pri3D: Can 3D Priors Help 2D Representation Learning?cited by 61
  36. 2021Weakly Supervised Learning of Rigid 3D Scene Flowcited by 60
  37. 2021Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR-based Perceptioncited by 57
  38. 2022PMP-Net++: Point Cloud Completion by Transformer-Enhanced Multi-step Point Moving Pathscited by 57

50 most cited of 318 All 318 (JSON)