Research map / 3D vision / Face and shape
EquiCAD: A Geometric Equivariant Neural Network for 3D Shape Classification
Yonghao Su, Yantao Gan, Junfeng Long, Caiyang Yu, Wenhao Zheng, Jiancheng Lv, Xianggen Liu
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 equivariant encoders to the CNN-only model (ablation) | EquiCAD · Parts | accuracy: 86.70% -> 91.66% | Improved |
| add equivariant encoders to the CNN-only model (ablation) | EquiCAD · SolidLetters | accuracy: 96.96% -> 97.64% | Improved |
| add CNN auxiliary branch to the equivariant-only model (ablation) | EquiCAD · Parts | accuracy: 88.03% -> 91.66% | Improved |
| add CNN auxiliary branch to the equivariant-only model (ablation) | EquiCAD · SolidLetters | accuracy: 95.27% -> 97.64% | Improved |
| add equivariant encoder to the curve branch (ablation) | EquiCAD · SolidLetters | accuracy: 96.12% -> 97.64% | Improved |
| add equivariant encoder to the surface branch (ablation) | EquiCAD · Parts | accuracy: 91.05% -> 91.66% | Improved |
| add CNN auxiliary to the curve branch (ablation) | EquiCAD · Parts | accuracy: 90.52% -> 91.66% | Improved |
| add CNN auxiliary to the surface branch (ablation) | EquiCAD · SolidLetters | accuracy: 95.40% -> 97.64% | Improved |
| replace DFGAT with EquiCAD under rotations | Parts · Parts (rotation robustness) | accuracy: 0.3615 -> 0.9159 | Improved |
| replace DFGAT with EquiCAD under reflection | Parts · Parts (reflection robustness) | accuracy: 0.1934 -> 0.9159 | Improved |
| replace UV-Net with EquiCAD | Machining Feature · Machining Feature | accuracy: 99.74 -> 99.87 | Improved |
| replace PointNet++ with EquiCAD | SolidLetters · SolidLetters | accuracy: 97.23 -> 97.64 | Improved |
Closest papers by meaning
- SO(3)-Equivariant Learning on CAD Boundary Representations (NeurIPS 2026 · similarity 0.92)
- A new perspective on building efficient and expressive 3D equivariant graph neural networks (NeurIPS 2023 · similarity 0.87)
- 3D Equivariant Graph Implicit Functions (ECCV 2022 · similarity 0.87)
- Learning Generalizable Shape Completion with SIM(3) Equivariance (NeurIPS 2025 · similarity 0.86)
- Geometry-Informed Neural Networks (ICML 2025 · similarity 0.84)
- Grounding Continuous Representations in Geometry: Equivariant Neural Fields (ICLR 2025 · 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.
