mlsanityData to 7 Oct 2026

Hierarchical Aggregation for 3D Instance Segmentation

arXiv 2108.02350 ↗First posted 5 Aug 2021cs.CVcited by 126

Not on the map. Why?

Abstract (arXiv)

Instance segmentation on point clouds is a fundamental task in 3D scene perception. In this work, we propose a concise clustering-based framework named HAIS, which makes full use of spatial relation of points and point sets. Considering clustering-based methods may result in over-segmentation or under-segmentation, we introduce the hierarchical aggregation to progressively generate instance proposals, i.e., point aggregation for preliminarily clustering points to sets and set aggregation for generating complete instances from sets. Once the complete 3D instances are obtained, a sub-network of intra-instance prediction is adopted for noisy points filtering and mask quality scoring. HAIS is fast (only 410ms per frame) and does not require non-maximum suppression. It ranks 1st on the ScanNet v2 benchmark, achieving the highest 69.9% AP50 and surpassing previous state-of-the-art (SOTA) methods by a large margin. Besides, the SOTA results on the S3DIS dataset validate the good generalization ability. Code will be available at https://github.com/hustvl/HAIS.

Key facts. Cited by 126 papers in the map (data through 7 Oct 2026). Builds on 42 of its 48 references that are in the map. First posted on arXiv 5 Aug 2021.

Builds on

42 of 48 references.

References · 42

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  2. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  3. 2017Mask R-CNNcited by 7,259
  4. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  5. 2016Feature Pyramid Networks for Object Detectioncited by 5,021
  6. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  7. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  8. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  9. 2015Fast R-CNNcited by 2,930
  10. 2018Path Aggregation Network for Instance Segmentationcited by 796
  11. 20173D Semantic Segmentation with Submanifold Sparse Convolutional Networkscited by 746
  12. 2018PartNet: A Large-scale Benchmark for Fine-grained and Hierarchical Part-level 3D Object Understandingcited by 521
30 more
  1. 2019Hybrid Task Cascade for Instance Segmentationcited by 479
  2. 2016R-FCN: Object Detection via Region-based Fully Convolutional Networkscited by 414
  3. 2016OctNet: Learning Deep 3D Representations at High Resolutionscited by 340
  4. 2020PointGroup: Dual-Set Point Grouping for 3D Instance Segmentationcited by 318
  5. 2019Mask Scoring R-CNNcited by 265
  6. 20183D-SIS: 3D Semantic Instance Segmentation of RGB-D Scanscited by 216
  7. 2017SEGCloud: Semantic Segmentation of 3D Point Cloudscited by 178
  8. 2020OccuSeg: Occupancy-aware 3D Instance Segmentationcited by 177
  9. 2018Acquisition of Localization Confidence for Accurate Object Detectioncited by 174
  10. 2017SGPN: Similarity Group Proposal Network for 3D Point Cloud Instance Segmentationcited by 168
  11. 2021Instances as Queriescited by 157
  12. 2019Learning Object Bounding Boxes for 3D Instance Segmentation on Point Cloudscited by 156
  13. 2018GSPN: Generative Shape Proposal Network for 3D Instance Segmentation in Point Cloudcited by 145
  14. 20203D-MPA: Multi Proposal Aggregation for 3D Semantic Instance Segmentationcited by 140
  15. 2016Fully Convolutional Instance-aware Semantic Segmentationcited by 122
  16. 2017Semantic Instance Segmentation with a Discriminative Loss Functioncited by 118
  17. 2019GS3D: An Efficient 3D Object Detection Framework for Autonomous Drivingcited by 106
  18. 2019PanopticFusion: Online Volumetric Semantic Mapping at the Level of Stuff and Thingscited by 97
  19. 2019Instance Segmentation by Jointly Optimizing Spatial Embeddings and Clustering Bandwidthcited by 88
  20. 2019Associatively Segmenting Instances and Semantics in Point Cloudscited by 86
  21. 20193D Instance Segmentation via Multi-Task Metric Learningcited by 79
  22. 2016Deep Watershed Transform for Instance Segmentationcited by 77
  23. 2020Boundary-preserving Mask R-CNNcited by 74
  24. 2019JSIS3D: Joint Semantic-Instance Segmentation of 3D Point Clouds with Multi-Task Pointwise Networks and Multi-Value Conditional Random Fieldscited by 62
  25. 2017Recurrent Pixel Embedding for Instance Groupingcited by 57
  26. 2017Semantic Instance Segmentation via Deep Metric Learningcited by 37
  27. 2016Instance-sensitive Fully Convolutional Networkscited by 36
  28. 2020Learning Gaussian Instance Segmentation in Point Cloudscited by 33
  29. 2019MASC: Multi-scale Affinity with Sparse Convolution for 3D Instance Segmentationcited by 25
  30. 2019EmbedMask: Embedding Coupling for One-stage Instance Segmentationcited by 15

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 4; 2022: 26; 2023: 38; 2024: 26; 2025: 20; 2026: 12.

126 papers.

Sort:

By area:

  1. 2023ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenescited by 747
  2. 2022Mask3D: Mask Transformer for 3D Semantic Instance Segmentationcited by 299
  3. 2022SoftGroup for 3D Instance Segmentation on Point Cloudscited by 197
  4. 2022GAPartNet: Cross-Category Domain-Generalizable Object Perception and Manipulation via Generalizable and Actionable Partscited by 135
  5. 2023Open3DIS: Open-Vocabulary 3D Instance Segmentation with 2D Mask Guidancecited by 123
  6. 2022Superpoint Transformer for 3D Scene Instance Segmentationcited by 115
  7. 2022EDA: Explicit Text-Decoupling and Dense Alignment for 3D Visual Groundingcited by 98
  8. 2023OneFormer3D: One Transformer for Unified Point Cloud Segmentationcited by 95
  9. 2022Compositional Human-Scene Interaction Synthesis with Semantic Controlcited by 82
  10. 2023Segment3D: Learning Fine-Grained Class-Agnostic 3D Segmentation without Manual Labelscited by 64
  11. 2023Efficient 3D Semantic Segmentation with Superpoint Transformercited by 56
  12. 2021D3Net: A Unified Speaker-Listener Architecture for 3D Dense Captioning and Visual Groundingcited by 52
38 more
  1. 2022STPLS3D: A Large-Scale Synthetic and Real Aerial Photogrammetry 3D Point Cloud Datasetcited by 52
  2. 2023ISBNet: a 3D Point Cloud Instance Segmentation Network with Instance-aware Sampling and Box-aware Dynamic Convolutioncited by 52
  3. 2023Mask-Attention-Free Transformer for 3D Instance Segmentationcited by 45
  4. 20223D Instances as 1D Kernelscited by 36
  5. 2023AGILE3D: Attention Guided Interactive Multi-object 3D Segmentationcited by 33
  6. 2023Top-Down Beats Bottom-Up in 3D Instance Segmentationcited by 31
  7. 2023UnScene3D: Unsupervised 3D Instance Segmentation for Indoor Scenescited by 31
  8. 2024Open-YOLO 3D: Towards Fast and Accurate Open-Vocabulary 3D Instance Segmentationcited by 30
  9. 2022MaskGroup: Hierarchical Point Grouping and Masking for 3D Instance Segmentationcited by 29
  10. 2024Reason3D: Searching and Reasoning 3D Segmentation via Large Language Modelcited by 27
  11. 2022Divide and Conquer: 3D Point Cloud Instance Segmentation With Point-Wise Binarizationcited by 26
  12. 2024ODIN: A Single Model for 2D and 3D Segmentationcited by 25
  13. 2023UrbanBIS: a Large-scale Benchmark for Fine-grained Urban Building Instance Segmentationcited by 24
  14. 2022Interactive Object Segmentation in 3D Point Cloudscited by 23
  15. 2022RaLiBEV: Radar and LiDAR BEV Fusion Learning for Anchor Box Free Object Detection Systemscited by 23
  16. 2023Spherical Mask: Coarse-to-Fine 3D Point Cloud Instance Segmentation with Spherical Representationcited by 23
  17. 2023Automated forest inventory: analysis of high-density airborne LiDAR point clouds with 3D deep learningcited by 23
  18. 2021Deep Instance Segmentation with Automotive Radar Detection Pointscited by 22
  19. 2022Point Scene Understanding via Disentangled Instance Mesh Reconstructioncited by 20
  20. 2023BUOL: A Bottom-Up Framework with Occupancy-aware Lifting for Panoptic 3D Scene Reconstruction From A Single Imagecited by 20
  21. 2024UniDet3D: Multi-dataset Indoor 3D Object Detectioncited by 20
  22. 2023EFEM: Equivariant Neural Field Expectation Maximization for 3D Object Segmentation Without Scene Supervisioncited by 18
  23. 2023Human-centric Scene Understanding for 3D Large-scale Scenarioscited by 17
  24. 2025TSP3D: Text-guided Sparse Voxel Pruning for Efficient 3D Visual Groundingcited by 17
  25. 2022PointInst3D: Segmenting 3D Instances by Pointscited by 16
  26. 2021Point Cloud Instance Segmentation with Semi-supervised Bounding-Box Miningcited by 15
  27. 2023Dense Object Grounding in 3D Scenescited by 15
  28. 2024A Unified Framework for 3D Scene Understandingcited by 15
  29. 2022Scalable SoftGroup for 3D Instance Segmentation on Point Cloudscited by 14
  30. 20223D-QueryIS: A Query-based Framework for 3D Instance Segmentationcited by 14
  31. 2023Point-GCC: Universal Self-supervised 3D Scene Pre-training via Geometry-Color Contrastcited by 14
  32. 2023TreeLearn: A deep learning method for segmenting individual trees from ground-based LiDAR forest point cloudscited by 14
  33. 2024Scalable 3D Panoptic Segmentation As Superpoint Graph Clusteringcited by 14
  34. 2022Learning Inter-Superpoint Affinity for Weakly Supervised 3D Instance Segmentationcited by 13
  35. 2024Segment Any 3D Object with Languagecited by 13
  36. 2022NeuralBF: Neural Bilateral Filtering for Top-down Instance Segmentation on Point Cloudscited by 12
  37. 2024Part2Object: Hierarchical Unsupervised 3D Instance Segmentationcited by 12
  38. 2024SGIFormer: Semantic-guided and Geometric-enhanced Interleaving Transformer for 3D Instance Segmentationcited by 12

50 most cited of 126 All 126 (JSON)