Research map / Point clouds / Segmentation and detection
GrowSP: Unsupervised Semantic Segmentation of 3D Point Clouds
Zihui Zhang, Bo Yang, Bing Wang, Bo Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| remove superpoint construction and progressive growing (ablation) | GrowSP · S3DIS Area-5 | mIoU: 44.5±1.1 -> 20.3±0.4 | Hurt |
| replace semantic primitive clustering with direct 12-class grouping (ablation) | GrowSP · S3DIS Area-5 | mIoU: 44.5±1.1 -> 25.4±1.0 | Hurt |
| remove PFH geometry features from semantic clustering (ablation) | GrowSP · S3DIS Area-5 | mIoU: 44.5±1.1 -> 38.9±0.9 | Hurt |
| use 25cm rather than 50cm initial voxels (ablation) | GrowSP · S3DIS Area-5 | mIoU: 44.5±1.1 -> 41.3±1.8 | Hurt |
| use 75cm rather than 50cm initial voxels (ablation) | GrowSP · S3DIS Area-5 | mIoU: 44.5±1.1 -> 43.2±0.7 | Hurt |
| set M1 to 160 (ablation) | GrowSP · S3DIS Area-5 | mIoU: 44.5±1.1 -> 43.3±1.3 | Hurt |
| set M1 to 120 (ablation) | GrowSP · S3DIS Area-5 | mIoU: 44.5±1.1 -> 41.3±3.2 | Hurt |
| set M1 to 40 (ablation) | GrowSP · S3DIS Area-5 | mIoU: 44.5±1.1 -> 43.1±2.0 | Hurt |
| set final superpoint count MT to 60 (ablation) | GrowSP · S3DIS Area-5 | mIoU: 44.5±1.1 -> 42.4±1.1 | Hurt |
| set final superpoint count MT to 40 (ablation) | GrowSP · S3DIS Area-5 | mIoU: 44.5±1.1 -> 43.0±0.8 | Hurt |
| set final superpoint count MT to 12 (ablation) | GrowSP · S3DIS Area-5 | mIoU: 44.5±1.1 -> 38.9±3.0 | Hurt |
| reduce superpoint count by 1 every 40 epochs (ablation) | GrowSP · S3DIS Area-5 | mIoU: 44.5±1.1 -> 42.2±2.4 | Hurt |
Later papers that compared with this method
4 later accepted papers report GrowSP as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.
- P-SLCR: Unsupervised Point Cloud Semantic Segmentation via Prototypes Structure Learning and Consistent Reasoning (AAAI 2026 · better on 12 of 13 shared results)
- PointGS: Semantic-Consistent Unsupervised 3D Point Cloud Segmentation with 3D Gaussian Splatting (CVPR 2026 · better on 3 of 3 shared results)
- GeoFree-CoSeg: Unsupervised Point Cloud-Image Cross-Modal Co-Segmentation Without Geometric Alignment (CVPR 2026 · better on 1 of 1 shared result)
- LogoSP: Local-global Grouping of Superpoints for Unsupervised Semantic Segmentation of 3D Point Clouds (CVPR 2025 · better on 5 of 7 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- GrowSP++: Growing Superpoints and Primitives for Unsupervised 3D Semantic Segmentation (TPAMI 2026 · similarity 0.97)
- LogoSP: Local-global Grouping of Superpoints for Unsupervised Semantic Segmentation of 3D Point Clouds (CVPR 2025 · similarity 0.93)
- PointDC: Unsupervised Semantic Segmentation of 3D Point Clouds via Cross-Modal Distillation and Super-Voxel Clustering (ICCV 2023 · similarity 0.92)
- SSPC-Net: Semi-supervised Semantic 3D Point Cloud Segmentation Network (AAAI 2021 · similarity 0.90)
- FreePoint: Unsupervised Point Cloud Instance Segmentation (CVPR 2024 · similarity 0.89)
- Unsupervised Semantic Segmentation Through Depth-Guided Feature Correlation and Sampling (CVPR 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
