Research map / 3D vision / Face and shape
Field Convolutions for Surface CNNs
Thomas W. Mitchel, Vladimir G. Kim, Michael Kazhdan
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 field convolutions (FC) instead of HSN | surface CNN · SHREC '11 classification | accuracy: 96.1% -> 99.2% | Improved |
| use field convolutions (FC) instead of DiffusionNet | surface CNN · SHREC '11 classification | accuracy: 98.9% -> 99.2% | Improved |
| use field convolutions (FC) instead of HSN | surface CNN · Human body segmentation (composite dataset) | accuracy: 91.1% -> 92.9% | Improved |
| use field convolutions (FC) instead of DiffusionNet | surface CNN · Human body segmentation (composite dataset) | accuracy: 91.5% -> 92.9% | Improved |
| use field convolutions (FC) instead of MeshWalker | surface CNN · Human body segmentation (composite dataset) | accuracy: 92.7% -> 92.9% | Improved |
| use field convolutions (FC) instead of HSN with half the memory footprint | surface CNN · parameters per convolution | parameters per convolution: 28 -> 33 | Mixed |
| use FCNet instead of HSN and ACSCNN with raw 3D coordinates input | dense correspondence · FAUST remeshed | correspondence percentage vs geodesic error | Improved |
| use FCNet instead of HSN and ACSCNN | feature matching · SHREC 2019 shape correspondence | precision-recall | Improved |
Closest papers by meaning
- FPNN: Field Probing Neural Networks for 3D Data (NeurIPS 2016 · similarity 0.88)
- Neural Convolutional Surfaces (CVPR 2022 · similarity 0.88)
- TextureNet: Consistent Local Parametrizations for Learning From High-Resolution Signals on Meshes (CVPR 2019 · similarity 0.87)
- PFCNN: Convolutional Neural Networks on 3D Surfaces Using Parallel Frames (CVPR 2020 · similarity 0.87)
- Learning on the Edge: Investigating Boundary Filters in CNNs (IJCV 2019 · similarity 0.87)
- Improving Neural Network Surface Processing with Principal Curvatures (NeurIPS 2024 · similarity 0.86)
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.
