mlsanityData to 7 Oct 2026

A Review of Panoptic Segmentation for Mobile Mapping Point Clouds

arXiv 2304.13980 ↗First posted 27 Apr 2023cs.CVcited by 8

Not on the map. Why?

Abstract (arXiv)

3D point cloud panoptic segmentation is the combined task to (i) assign each point to a semantic class and (ii) separate the points in each class into object instances. Recently there has been an increased interest in such comprehensive 3D scene understanding, building on the rapid advances of semantic segmentation due to the advent of deep 3D neural networks. Yet, to date there is very little work about panoptic segmentation of outdoor mobile-mapping data, and no systematic comparisons. The present paper tries to close that gap. It reviews the building blocks needed to assemble a panoptic segmentation pipeline and the related literature. Moreover, a modular pipeline is set up to perform comprehensive, systematic experiments to assess the state of panoptic segmentation in the context of street mapping. As a byproduct, we also provide the first public dataset for that task, by extending the NPM3D dataset to include instance labels. That dataset and our source code are publicly available. We discuss which adaptations are need to adapt current panoptic segmentation methods to outdoor scenes and large objects. Our study finds that for mobile mapping data, KPConv performs best but is slower, while PointNet++ is fastest but performs significantly worse. Sparse CNNs are in between. Regardless of the backbone, Instance segmentation by clustering embedding features is better than using shifted coordinates.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 70 of its 102 references that are in the map. First posted on arXiv 27 Apr 2023.

Builds on

70 of 102 references.

References · 70

  1. 2017Mask R-CNNcited by 7,259
  2. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  3. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  4. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  5. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  6. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  7. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  8. 2019KPConv: Flexible and Deformable Convolution for Point Cloudscited by 1,202
  9. 20194D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networkscited by 1,201
  10. 2018Panoptic Segmentationcited by 796
  11. 20173D Semantic Segmentation with Submanifold Sparse Convolutional Networkscited by 746
  12. 2019Deep Learning for 3D Point Clouds: A Surveycited by 557
38 more
  1. 2018PartNet: A Large-scale Benchmark for Fine-grained and Hierarchical Part-level 3D Object Understandingcited by 521
  2. 2016Semantic Scene Completion from a Single Depth Imagecited by 468
  3. 2020Searching Efficient 3D Architectures with Sparse Point-Voxel Convolutioncited by 422
  4. 2017Large-scale Point Cloud Semantic Segmentation with Superpoint Graphscited by 384
  5. 2016OctNet: Learning Deep 3D Representations at High Resolutionscited by 340
  6. 2020PointGroup: Dual-Set Point Grouping for 3D Instance Segmentationcited by 318
  7. 2017SqueezeSeg: Convolutional Neural Nets with Recurrent CRF for Real-Time Road-Object Segmentation from 3D LiDAR Point Cloudcited by 231
  8. 2018SqueezeSegV2: Improved Model Structure and Unsupervised Domain Adaptation for Road-Object Segmentation from a LiDAR Point Cloudcited by 228
  9. 2020SalsaNext: Fast, Uncertainty-aware Semantic Segmentation of LiDAR Point Clouds for Autonomous Drivingcited by 226
  10. 2022SoftGroup for 3D Instance Segmentation on Point Cloudscited by 197
  11. 2020OccuSeg: Occupancy-aware 3D Instance Segmentationcited by 177
  12. 2018Tangent Convolutions for Dense Prediction in 3Dcited by 173
  13. 2017SGPN: Similarity Group Proposal Network for 3D Point Cloud Instance Segmentationcited by 168
  14. 2021Panoptic nuScenes: A Large-Scale Benchmark for LiDAR Panoptic Segmentation and Trackingcited by 165
  15. 2017Semantic3D.net: A new Large-scale Point Cloud Classification Benchmarkcited by 158
  16. 2018SPLATNet: Sparse Lattice Networks for Point Cloud Processingcited by 158
  17. 2019Learning Object Bounding Boxes for 3D Instance Segmentation on Point Cloudscited by 156
  18. 2018GSPN: Generative Shape Proposal Network for 3D Instance Segmentation in Point Cloudcited by 145
  19. 20203D-MPA: Multi Proposal Aggregation for 3D Semantic Instance Segmentationcited by 140
  20. 2017Pointwise Convolutional Neural Networkscited by 126
  21. 2021Hierarchical Aggregation for 3D Instance Segmentationcited by 126
  22. 2017Semantic Instance Segmentation with a Discriminative Loss Functioncited by 118
  23. 20183DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentationcited by 118
  24. 2022Superpoint Transformer for 3D Scene Instance Segmentationcited by 115
  25. 2020Deep Learning for LiDAR Point Clouds in Autonomous Driving: A Reviewcited by 114
  26. 2019Seamless Scene Segmentationcited by 112
  27. 2020Cylinder3D: An Effective 3D Framework for Driving-scene LiDAR Semantic Segmentationcited by 112
  28. 2021Panoptic-PolarNet: Proposal-free LiDAR Point Cloud Panoptic Segmentationcited by 107
  29. 2021Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networkscited by 98
  30. 2019Instance Segmentation by Jointly Optimizing Spatial Embeddings and Clustering Bandwidthcited by 88
  31. 2019Associatively Segmenting Instances and Semantics in Point Cloudscited by 86
  32. 2020LiDAR-based Panoptic Segmentation via Dynamic Shifting Networkcited by 84
  33. 2017Deep Projective 3D Semantic Segmentationcited by 81
  34. 20193D Instance Segmentation via Multi-Task Metric Learningcited by 79
  35. 2019Multi-view PointNet for 3D Scene Understandingcited by 76
  36. 2020DyCo3D: Robust Instance Segmentation of 3D Point Clouds through Dynamic Convolutioncited by 74
  37. 2017Paris-Lille-3D: a large and high-quality ground truth urban point cloud dataset for automatic segmentation and classificationcited by 65
  38. 20163D Shape Segmentation with Projective Convolutional Networkscited by 63

50 most cited of 70 All 70 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 2; 2025: 2; 2026: 2.

8 papers.

  1. 2023Automated forest inventory: analysis of high-density airborne LiDAR point clouds with 3D deep learningcited by 23
  2. 2024Scalable 3D Panoptic Segmentation As Superpoint Graph Clusteringcited by 14
  3. 2023Towards accurate instance segmentation in large-scale LiDAR point cloudscited by 8
  4. 2025IPFormer: Visual 3D Panoptic Scene Completion with Context-Adaptive Instance Proposalscited by 3
  5. 2025Benchmarking individual tree segmentation using multispectral airborne laser scanning data: the FGI-EMIT datasetcited by 2
  6. 2024Horticultural Temporal Fruit Monitoring via 3D Instance Segmentation and Re-Identification using Colored Point Cloudscited by 1
  7. 2026In-Field 3D Wheat Head Instance Segmentation From TLS Point Clouds Using Deep Learning Without Manual Labelscited by 0
  8. 2026LiPS: Lightweight Panoptic Segmentation for Resource-Constrained Roboticscited by 0