Research map / Point clouds / Segmentation and detection
Learning Superpoint Graph Cut for 3D Instance Segmentation
Le Hui, Linghua Tang, Yaqi Shen, Jin Xie, Jian Yang
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 |
|---|---|---|---|
| add coordinate-space edge embedding (ablation) | GraphCut · ScanNet v2 val | AP: 50.1 -> 50.9 | Improved |
| add feature-space edge embedding (ablation) | GraphCut · ScanNet v2 val | AP: 50.1 -> 51.5 | Improved |
| use both coordinate and feature edge embeddings (ablation) | GraphCut · ScanNet v2 val | AP: 50.1 -> 52.2 | Improved |
| add area constraint loss (ablation) | GraphCut · ScanNet v2 val | AP: 47.8 -> 51.7 | Improved |
| add feature loss constraint (ablation) | GraphCut · ScanNet v2 val | AP: 47.8 -> 51.2 | Improved |
| use both area and feature losses (ablation) | GraphCut · ScanNet v2 val | AP: 47.8 -> 52.2 | Improved |
| replace MLP+max-pool with bilateral graph attention (ablation) | GraphCut · ScanNet v2 val | AP: 49.9 -> 52.2 | Improved |
| k-NN size k=2 vs k=4 (ablation) | GraphCut · ScanNet v2 val | AP: 52.2 -> 52.0 | Hurt |
| k-NN size k=8 vs k=4 (ablation) | GraphCut · ScanNet v2 val | AP: 52.2 -> 51.4 | Hurt |
| k-NN size k=16 vs k=4 (ablation) | GraphCut · ScanNet v2 val | AP: 52.2 -> 50.8 | Hurt |
| GraphCut vs second-best method SSTNet | GraphCut · ScanNet v2 hidden test | AP: 50.6 -> 55.2 | Improved |
| GraphCut vs SoftGroup | GraphCut · S3DIS Area 5 | AP: 51.6 -> 54.1 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- 3D-MPA — 3D-MPA: Multi-Proposal Aggregation for 3D Semantic Instance Segmentation (CVPR 2020) · better on 4 of 4 shared results
- PointGroup — PointGroup: Dual-Set Point Grouping for 3D Instance Segmentation (CVPR 2020) · better on 4 of 4 shared results
- GSPN — GSPN: Generative Shape Proposal Network for 3D Instance Segmentation in Point Cloud (CVPR 2019) · better on 4 of 4 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
- SAM-guided Graph Cut for 3D Instance Segmentation (ECCV 2024 · similarity 0.94)
- Instance Segmentation in 3D Scenes Using Semantic Superpoint Tree Networks (ICCV 2021 · similarity 0.90)
- Edge-Aware 3D Instance Segmentation Network with Intelligent Semantic Prior (CVPR 2024 · similarity 0.88)
- FreePoint: Unsupervised Point Cloud Instance Segmentation (CVPR 2024 · similarity 0.88)
- Relation3D : Enhancing Relation Modeling for Point Cloud Instance Segmentation (CVPR 2025 · similarity 0.87)
- SA3DIP: Segment Any 3D Instance with Potential 3D Priors (NeurIPS 2024 · similarity 0.87)
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.
