Research map / Point clouds / Segmentation and detection
JSIS3D: Joint Semantic-Instance Segmentation of 3D Point Clouds With Multi-Task Pointwise Networks and Multi-Value Conditional Random Fields
Quang-Hieu Pham, Thanh Nguyen, Binh-Son Hua, Gemma Roig, Sai-Kit Yeung
Check 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 semantic pairwise potentials to semantic unary potentials (ablation) | MV-CRF · S3DIS | mAcc: 86.7 -> 86.9 | Improved |
| Jointly optimize semantic and instance labels with the full MV-CRF (ablation) | MV-CRF · S3DIS | mAcc: 86.9 -> 87.4 | Improved |
| Add instance pairwise potentials to instance unary potentials (ablation) | MV-CRF · S3DIS | mAP@0.5: 24.9 -> 27.4 | Improved |
| Jointly optimize semantic and instance labels with the full MV-CRF (ablation) | MV-CRF · S3DIS | mAP@0.5: 27.4 -> 36.3 | Improved |
| Add the instance-embedding branch to the pointwise network (ablation) | MT-PNet · S3DIS semantic segmentation | semantic segmentation performance | Improved |
| Disable the instance-embedding branch (ablation) | MT-PNet · S3DIS semantic segmentation | semantic segmentation performance | Hurt |
| Use a lower mean-shift bandwidth than the embedding push margin (ablation) | MT-PNet instance clustering · S3DIS and SceneNN | segmentation performance | Hurt |
Closest papers by meaning
- Associatively Segmenting Instances and Semantics in Point Clouds (CVPR 2019 · similarity 0.92)
- JSNet: Joint Instance and Semantic Segmentation of 3D Point Clouds (AAAI 2020 · similarity 0.92)
- 3D Instance Segmentation via Enhanced Spatial and Semantic Supervision (ICCV 2023 · similarity 0.90)
- Weakly Supervised 3D Scene Segmentation with Region-Level Boundary Awareness and Instance Discrimination (ECCV 2022 · similarity 0.90)
- Relation3D : Enhancing Relation Modeling for Point Cloud Instance Segmentation (CVPR 2025 · similarity 0.89)
- Segment Any 3D Object with Language (ICLR 2025 · similarity 0.89)
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.
