Research map / Point clouds / Segmentation and detection
P3-SAM: Native 3D Part Segmentation
CHANGFENG MA, YANG LI, Xinhao Yan, Jiachen Xu, Yunhan Yang, Chunshi Wang, Zibo Zhao, Yanwen Guo, Zhuo Chen, Chunchao Guo
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 |
|---|---|---|---|
| use P3-SAM instead of PartField (best baseline) for automatic segmentation | PartField · PartObj-Tiny (seg. w/o connectivity) | IoU: 53.93 -> 59.88 | Improved |
| use P3-SAM instead of PartField with connectivity | PartField · PartObj-Tiny (seg. w/ connectivity) | IoU: 79.18 -> 81.14 | Improved |
| use P3-SAM instead of Point-SAM for interactive segmentation | Point-SAM · PartObj-Tiny (interactive) | IoU: 27.91 -> 51.23 | Improved |
| use P3-SAM on watertight meshes instead of PartField | PartField · PartObj-Tiny-WT (seg. w/o connectivity) | IoU: 51.54 -> 58.10 | Improved |
| use P3-SAM on point clouds instead of PartField | PartField · PartNetE (seg. w/o connectivity) | IoU: 59.1 -> 65.39 | Improved |
| use P3-SAM instead of Point-SAM on point clouds (interactive) | Point-SAM · PartNetE (interactive) | IoU: 45.85 -> 63.48 | Improved |
| add second stage with global feature (stage 2 only vs stage 1 only) (ablation) | P3-SAM · Own test set (ablation) | mIoU: 0.4265 -> 0.6647 | Improved |
| use three heads plus two stages instead of single head stage 1 (ablation) | P3-SAM · Own test set (ablation) | mIoU: 0.2801 -> 0.7464 | Improved |
| combine stage 1 and stage 2 instead of stage 2 only (ablation) | P3-SAM · Own test set (ablation) | mIoU: 0.6647 -> 0.7464 | Improved |
| add data augmentation (ablation) | P3-SAM · Own test set (ablation) | mIoU: 0.7464 -> 0.7906 | Improved |
| remove NMS and flood fill from automatic segmentation (ablation) | P3-SAM · Qualitative (automatic segmentation) | qualitative | Improved |
Later papers that compared with this method
1 later accepted paper reports P3-SAM as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.
- S$^2$AM3D: Scale-controllable Part Segmentation of 3D Point Clouds (CVPR 2026 · better on 0 of 1 shared result)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Point-SAM — Point-SAM: Promptable 3D Segmentation Model for Point Clouds (ICLR 2025) · better on 3 of 3 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
- Point-SAM: Promptable 3D Segmentation Model for Point Clouds (ICLR 2025 · similarity 0.89)
- SegGraph: Leveraging Graphs of SAM Segments for Few-Shot 3D Part Segmentation (NeurIPS 2025 · similarity 0.88)
- AGILE3D: Attention Guided Interactive Multi-object 3D Segmentation (ICLR 2024 · similarity 0.87)
- PartNeXt: A Next-Generation Dataset for Fine-Grained and Hierarchical 3D Part Understanding (NeurIPS 2025 · similarity 0.86)
- SAI3D: Segment Any Instance in 3D Scenes (CVPR 2024 · similarity 0.86)
- SAM2Object: Consolidating View Consistency via SAM2 for Zero-Shot 3D Instance Segmentation (CVPR 2025 · similarity 0.85)
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.
