Research map / 3D vision / Gaussian splatting
Pointmap Association and Piecewise-Plane Constraint for Consistent and Compact 3D Gaussian Segmentation Field
Wenhao Hu, Wenhao Chai, Shengyu Hao, Xiaotong Cui, Xuexiang Wen, Jenq–Neng Hwang, Gaoang Wang
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 pointmap fusion (ablation) | Gaussian Grouping baseline · ScanNet | mIoU multi-view: 44.37 -> 60.18 | Improved |
| add pointmap fusion (ablation) | Gaussian Grouping baseline · ScanNet | mIoU single-view: 61.54 -> 63.38 | Improved |
| add pointmap fusion (ablation) | Gaussian Grouping baseline · ScanNet | mIoU 3D: 52.04 -> 57.12 | Improved |
| add plane regularization (ablation) | CCGS with pointmap fusion · ScanNet | mIoU 3D: 57.12 -> 59.56 | Improved |
| add plane regularization (ablation) | CCGS with pointmap fusion · ScanNet | Chamfer Distance: 0.480 -> 0.469 | Improved |
| add split projection (ablation) | CCGS with PF and PR · ScanNet | mIoU 3D: 59.56 -> 63.21 | Improved |
| add split projection (ablation) | CCGS with PF and PR · ScanNet | Chamfer Distance: 0.469 -> 0.451 | Improved |
| add split projection (ablation) | CCGS with PF and PR · ScanNet | mIoU single-view: 63.83 -> 63.92 | Improved |
| use CCGS instead of Gaussian Grouping | Gaussian Grouping · Replica | mIoU multi-view: 47.16 -> 62.31 | Improved |
| use CCGS instead of Gaussian Grouping | Gaussian Grouping · ScanNet 3D segmentation | mIoU 3D: 52.04 -> 63.21 | Improved |
| use too few (4) nearest neighbors for the piecewise plane (ablation) | CCGS · ScanNet | Chamfer Distance | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- Gaussian Grouping — Gaussian Grouping: Segment and Edit Anything in 3D Scenes (ECCV 2024) · better on 9 of 9 shared results
- SAGA — SAGA: Stochastic Whole-Body Grasping with Contact (ECCV 2022) · 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
- PointGS: Semantic-Consistent Unsupervised 3D Point Cloud Segmentation with 3D Gaussian Splatting (CVPR 2026 · similarity 0.87)
- G2P: Gaussian-to-Point Attribute Alignment for Boundary-Aware 3D Segmentation (ECCV 2026 · similarity 0.86)
- Bootstraping Clustering of Gaussians for View-consistent 3D Scene Understanding (AAAI 2025 · similarity 0.86)
- AGILE3D: Attention Guided Interactive Multi-object 3D Segmentation (ICLR 2024 · similarity 0.85)
- Any3DIS: Class-Agnostic 3D Instance Segmentation by 2D Mask Tracking (CVPR 2025 · similarity 0.84)
- Open3DIS: Open-Vocabulary 3D Instance Segmentation with 2D Mask Guidance (CVPR 2024 · similarity 0.84)
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.
