Research map / Point clouds / Point cloud classification
Unified Fourier-based Kernel and Nonlinearity Design for Equivariant Networks on Homogeneous Spaces
Yinshuang Xu, Jiahui Lei, Edgar Dobriban, Kostas Daniilidis
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 proposed equivariant activation to TFN | TFN · QM9 missing atom prediction, 19 atoms | accuracy (%): 93.9 -> 98.0 | Improved |
| Add the proposed equivariant activation to TFN | TFN · QM9 missing atom prediction, 23 atoms | Distance MAE (Å): 0.13 -> 0.10 | Improved |
| Add the proposed equivariant activation to TFN | TFN · QM9 missing atom prediction, 25-29 atoms | accuracy (%): 97.3 -> 98.3 | Improved |
| Use the proposed model rather than SWSCNN | SVMNIST · NR/R | MSE × 10^3: 4.3 -> 3.8 | Improved |
| Use the proposed model rather than Planar | SVMNIST · NR/NR | MSE × 10^3: 0.3 -> 2.9 | Hurt |
| Use the proposed model rather than VN-DGCNN | ModelNet40 classification · SO(3)/SO(3) | accuracy (%): 90.2 -> 89.3 | Hurt |
Closest papers by meaning
- Lie Group Decompositions for Equivariant Neural Networks (ICLR 2024 · similarity 0.88)
- Neural Fourier Transform: A General Approach to Equivariant Representation Learning (ICLR 2024 · similarity 0.88)
- Equivariant Neural Tangent Kernels (ICML 2025 · similarity 0.88)
- Fast, Expressive $\mathrm{SE}(n)$ Equivariant Networks through Weight-Sharing in Position-Orientation Space (ICLR 2024 · similarity 0.88)
- On the Fourier analysis in the SO(3) space : the EquiLoPO Network (ICLR 2025 · similarity 0.88)
- A Characterization Theorem for Equivariant Networks with Point-wise Activations (ICLR 2024 · similarity 0.87)
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.
