Papers › A Review of Panoptic Segmentation for Mobile Mapping Point Clouds

A Review of Panoptic Segmentation for Mobile Mapping Point Clouds

27 Apr 2023arXiv:2304.13980archive 2025-07-28

Binbin Xiang, Yuanwen Yue, Torben Peters, Konrad Schindler

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.

PaperPDFCode

Code

bxiang233/panopticsegformobilemappingpointclouds officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Instance SegmentationPanoptic SegmentationScene UnderstandingSegmentationSemantic Segmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Panoptic Segmentation S3DIS Area5 PointGroup (Xiang 2023) PQ 42.3 #2 of 5 Archive leaderboard report
Panoptic Segmentation S3DIS Area5 PointGroup (Xiang 2023) Params (M) 7.7 #2 of 5 Archive leaderboard report
Panoptic Segmentation S3DIS Area5 PointGroup (Xiang 2023) RQ 52.0 #2 of 5 Archive leaderboard report
Panoptic Segmentation S3DIS Area5 PointGroup (Xiang 2023) SQ 74.7 #2 of 5 Archive leaderboard report
Panoptic Segmentation S3DIS Area5 KPConv (Xiang 2023) PQ 41.8 #3 of 5 Archive leaderboard report
Panoptic Segmentation S3DIS Area5 KPConv (Xiang 2023) Params (M) 14.2 #3 of 5 Archive leaderboard report
Panoptic Segmentation S3DIS Area5 KPConv (Xiang 2023) RQ 51.5 #3 of 5 Archive leaderboard report
Panoptic Segmentation S3DIS Area5 KPConv (Xiang 2023) SQ 74.7 #3 of 5 Archive leaderboard report
Panoptic Segmentation S3DIS Area5 MinkowskiNet (Xiang 2023) PQ 39.2 #4 of 5 Archive leaderboard report
Panoptic Segmentation S3DIS Area5 MinkowskiNet (Xiang 2023) Params (M) 38.0 #4 of 5 Archive leaderboard report
Panoptic Segmentation S3DIS Area5 MinkowskiNet (Xiang 2023) RQ 48.0 #4 of 5 Archive leaderboard report
Panoptic Segmentation S3DIS Area5 MinkowskiNet (Xiang 2023) SQ 74.9 #4 of 5 Archive leaderboard report
Panoptic Segmentation S3DIS Area5 PointNet++ (Xiang 2023) PQ 24.6 #5 of 5 Archive leaderboard report
Panoptic Segmentation S3DIS Area5 PointNet++ (Xiang 2023) Params (M) 3.1 #5 of 5 Archive leaderboard report
Panoptic Segmentation S3DIS Area5 PointNet++ (Xiang 2023) RQ 32.6 #5 of 5 Archive leaderboard report
Panoptic Segmentation S3DIS Area5 PointNet++ (Xiang 2023) SQ 68.2 #5 of 5 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections