Research map / Point clouds / Segmentation and detection
JSNet: Joint Instance and Semantic Segmentation of 3D Point Clouds
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| jointly predict instance and semantic segmentation with JSNet rather than use ASIS | S3DIS Area 5 · S3DIS Area 5 | mPrec: 55.3 -> 62.1 | Helped |
| jointly predict instance and semantic segmentation with JSNet rather than use ASIS | S3DIS 6 fold · S3DIS 6 fold | mCov: 51.2 -> 54.1 | Helped |
| use JSNet rather than PointNet++ | ShapeNet · ShapeNet | mIoU: 84.9 -> 85.8 | Helped |
| replace PointNet++ base network with the PointNet++/PointConv backbone (ablation) | JSNet · S3DIS Area 5 | mPrec: 52.3 -> 55.9 | Helped |
| add PCFF to the backbone (ablation) | JSNet backbone · S3DIS Area 5 | mPrec: 55.9 -> 58.6 | Helped |
| add the instance fusion branch of JISS (ablation) | JSNet backbone · S3DIS Area 5 | mPrec: 55.9 -> 56.9 | Helped |
| add the semantic fusion branch of JISS (ablation) | JSNet backbone · S3DIS Area 5 | mPrec: 55.9 -> 57.2 | Helped |
| add both instance and semantic fusion branches (ablation) | JSNet backbone · S3DIS Area 5 | mPrec: 55.9 -> 58.6 | Helped |
| add PCFF to the backbone with both JISS branches (ablation) | JSNet · S3DIS Area 5 | mPrec: 58.6 -> 57.6 | Hurt |
| use early stopping with the full model (ablation) | JSNet · S3DIS Area 5 | mPrec: 57.6 -> 58.7 | Helped |
| train with random samples instead of fixed samples (ablation) | JSNet · S3DIS Area 5 | mPrec: 57.6 -> 62.1 | Helped |
| add early stopping to random-sample training (ablation) | JSNet · S3DIS Area 5 | mPrec: 62.1 -> 62.9 | Helped |
Closest papers by meaning
- OneFormer3D: One Transformer for Unified Point Cloud Segmentation (CVPR 2024 · similarity 0.87)
- Relation3D : Enhancing Relation Modeling for Point Cloud Instance Segmentation (CVPR 2025 · similarity 0.86)
- Insightful Instance Features for 3D Instance Segmentation (CVPR 2025 · similarity 0.86)
- Open3DIS: Open-Vocabulary 3D Instance Segmentation with 2D Mask Guidance (CVPR 2024 · similarity 0.86)
- Edge-Aware 3D Instance Segmentation Network with Intelligent Semantic Prior (CVPR 2024 · similarity 0.86)
- FreePoint: Unsupervised Point Cloud Instance Segmentation (CVPR 2024 · similarity 0.86)
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.