Research map / Point clouds / Shape and surface completion
CSG-Stump: A Learning Friendly CSG-Like Representation for Interpretable Shape Parsing
Daxuan Ren, Jianmin Zheng, Jianfei Cai, Jiatong Li, Haiyong Jiang, Zhongang Cai, Junzhe Zhang, Liang Pan, Mingyuan Zhang, Haiyu Zhao, Shuai Yi
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 |
|---|---|---|---|
| use CSG-Stump Net instead of UCSG-Net | CSG-Stump Net · ShapeNet | L2 Chamfer Distance x1000: 2.085 -> 1.505 | Improved |
| increase available primitives from 32 to 256 (ablation) | CSG-Stump Net · ShapeNet Airplane | Chamfer Distance: 1.44 -> 1.22 | Improved |
| increase available primitives from 64 to 128 (ablation) | CSG-Stump Net · ShapeNet Airplane | Chamfer Distance: 1.40 -> 1.28 | Improved |
| increase intersection nodes from 32 to 256 (ablation) | CSG-Stump Net · ShapeNet Airplane | Chamfer Distance: 2.26 -> 1.22 | Improved |
| increase intersection nodes from 64 to 128 (ablation) | CSG-Stump Net · ShapeNet Airplane | Chamfer Distance: 2.28 -> 1.37 | Improved |
| go from 64 to 32 intersection nodes (ablation) | CSG-Stump Net · ShapeNet Airplane | Chamfer Distance: 2.28 -> 2.26 | No clear effect |
| add cylinder primitives to box-only (ablation) | CSG-Stump Net · ShapeNet Airplane | Chamfer Distance: 1.27 -> 1.24 | Improved |
| add cone and sphere primitives to box and cylinder (ablation) | CSG-Stump Net · ShapeNet Airplane | Chamfer Distance: 1.24 -> 1.22 | Improved |
| learn a dynamic complement weight instead of fixing it to zero | CSG-Stump Net · ShapeNet | Chamfer Distance: slightly better performance | Improved |
| solve connection matrices with binary programming on complex ShapeNet shapes | Gurobi solver · ShapeNet complex shapes | optimality | Hurt |
Closest papers by meaning
- Neural Shape Parsers for Constructive Solid Geometry (TPAMI 2020 · similarity 0.89)
- D$^2$CSG: Unsupervised Learning of Compact CSG Trees with Dual Complements and Dropouts (NeurIPS 2023 · similarity 0.88)
- UCSG-NET- Unsupervised Discovering of Constructive Solid Geometry Tree (NeurIPS 2020 · similarity 0.87)
- CSGNet: Neural Shape Parser for Constructive Solid Geometry (CVPR 2018 · similarity 0.87)
- PatchNets: Patch-Based Generalizable Deep Implicit 3D Shape Representations (ECCV 2020 · similarity 0.87)
- Interpretable3D: An Ad-Hoc Interpretable Classifier for 3D Point Clouds (AAAI 2024 · similarity 0.85)
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.
