Research map / Point clouds / Segmentation and detection
JSENet: Joint Semantic Segmentation and Edge Detection Network for 3D Point Clouds
Zeyu Hu, Mingmin Zhen, Xuyang Bai, Hongbo Fu, Chiew-lan Tai
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 the edge stream with naive combination to the segmentation stream (ablation) | JSENet segmentation stream · S3DIS Area-5 | mIoU (%): 64.7 -> 66.2 | Improved |
| add the joint refinement module (ablation) | JSENet segmentation · S3DIS Area-5 | mIoU (%): 66.2 -> 67.7 | Improved |
| add the joint refinement module (ablation) | JSENet edge detection · S3DIS Area-5 | mMF (ODS) (%): 30.5 -> 31.0 | Improved |
| naively combine segmentation stream with edge stream (ablation) | JSENet edge detection · S3DIS Area-5 | mMF (ODS) (%): 30.2 -> 30.5 | Improved |
| add enhanced feature extraction to the edge stream (ablation) | edge stream · S3DIS Area-5 | mMF (ODS) (%): 29.4 -> 29.9 | Improved |
| add hierarchical supervision to the edge stream (ablation) | edge stream · S3DIS Area-5 | mMF (ODS) (%): 29.9 -> 30.2 | Improved |
| use L_bce for first three layers and L_seg for last two instead of L_bce for all five (ablation) | hierarchical supervision · S3DIS Area-5 | mMF (ODS) (%): 30.1 -> 30.2 | Improved |
| add the dual semantic edge loss (ablation) | JSENet · S3DIS Area-5 | F-score (%): 22.7 -> 23.1 | Improved |
| JSENet joint training versus KPConv alone | KPConv rigid · S3DIS Area-5 | mIoU (%): 65.4 -> 67.7 | Improved |
| use JSENet instead of KPConv for edge detection | edge detection · ScanNet | mMF (ODS) (%): 34.8 -> 37.3 | Improved |
| use extended CASENet architecture instead of KPConv for 3D edge detection | edge detection · S3DIS Area-5 | mMF (ODS) (%): 29.4 -> 27.1 | Hurt |
Later papers that compared with this method
2 later accepted papers report JSENet as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- CGA-Net: Category Guided Aggregation for Point Cloud Semantic Segmentation (CVPR 2021 · better on 1 of 1 shared result)
- Cloud Transformers: A Universal Approach to Point Cloud Processing Tasks (ICCV 2021 · better on 1 of 1 shared result)
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.
- PointConv — PointConv: Deep Convolutional Networks on 3D Point Clouds (CVPR 2019) · better on 2 of 2 shared results
- MVPNet — MVPNet: Multi-View Point Regression Networks for 3D Object Reconstruction from A Single Image (AAAI 2019) · better on 1 of 1 shared result
- PointCNN — PointCNN: Convolution On X-Transformed Points (NeurIPS 2018) · better on 1 of 1 shared result
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
- BFANet: Revisiting 3D Semantic Segmentation with Boundary Feature Analysis (CVPR 2025 · similarity 0.88)
- JSIS3D: Joint Semantic-Instance Segmentation of 3D Point Clouds With Multi-Task Pointwise Networks and Multi-Value Conditional Random Fields (CVPR 2019 · similarity 0.88)
- Hierarchical Point-Edge Interaction Network for Point Cloud Semantic Segmentation (ICCV 2019 · similarity 0.87)
- JSNet: Joint Instance and Semantic Segmentation of 3D Point Clouds (AAAI 2020 · similarity 0.87)
- LogoSP: Local-global Grouping of Superpoints for Unsupervised Semantic Segmentation of 3D Point Clouds (CVPR 2025 · similarity 0.87)
- Edge-Aware 3D Instance Segmentation Network with Intelligent Semantic Prior (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.
