Research map / Architectures and efficiency / CNN design
Shape correspondence using anisotropic Chebyshev spectral CNNs
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 anisotropic convolution with learned Chebyshev spectral filters | Shape correspondence on FAUST · FAUST test | accuracy (r=0): 93.06 % -> 98.06 % | Helped |
| Use ACSCNN instead of SplineCNN | Shape correspondence on FAUST with SHOT input · FAUST test | accuracy (r=0): 99.12 % -> 98.06 % | Hurt |
| Use constant vector 1 instead of SHOT input | ACSCNN · FAUST test | accuracy (r=0): 98.06 % -> 98.98 % | Helped |
| Add PMF refinement | ACSCNN with constant vector 1 input · FAUST test | accuracy (r=0): 98.98 % -> 99.56 % | Helped |
| Use ACSCNN instead of ACNN | Shape correspondence · SHREC'16 Topology | accuracy (r=0): 62.8% -> 90.2% | Helped |
| Increase Chebyshev polynomial order (ablation) | ACSCNN · FAUST | accuracy (r=0) | Helped |
| Increase the number of rotation angles (ablation) | ACSCNN · FAUST | accuracy (r=0) | Helped |
| Increase the number of rotation angles or Chebyshev order (ablation) | ACSCNN · FAUST | computational cost | Hurt |
Closest papers by meaning
- Learning shape correspondence with anisotropic convolutional neural networks (NeurIPS 2016 · similarity 0.89)
- From Feature Learning to Spectral Basis Learning: A Unifying and Flexible Framework for Efficient and Robust Shape Matching (CVPR 2026 · similarity 0.88)
- Convolutional Neural Networks on Non-uniform Geometrical Signals Using Euclidean Spectral Transformation (ICLR 2019 · similarity 0.87)
- Learned Binary Spectral Shape Descriptor for 3D Shape Correspondence (CVPR 2016 · similarity 0.87)
- Learning Multi-resolution Functional Maps with Spectral Attention for Robust Shape Matching (NeurIPS 2022 · similarity 0.86)
- Spectral Meets Spatial: Harmonising 3D Shape Matching and Interpolation (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.