Research map / Point clouds / Shape and surface completion
Convolutional Neural Networks on Non-uniform Geometrical Signals Using Euclidean Spectral Transformation
Read the paperCheck 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 NUFT instead of binary pixels | MNIST polygon classification · MNIST | classification accuracy | Helped |
| Use NUFT instead of distance functions | MNIST polygon classification · MNIST | classification accuracy | Helped |
| Use NUFT instead of downsampled raw pixels at low resolutions | MNIST polygon classification · MNIST | classification accuracy | Helped |
| Use NUFT-S or NUFT-V with a 3D DLA backbone | 3D shape retrieval · SHREC17 | retrieval performance | Helped |
| Compare NUFT retrieval with pretrained 2D models | 3D shape retrieval · SHREC17 | retrieval performance | No clear effect |
| Use NUFT-volume instead of NUFT-surface at the same resolution | 3D shape retrieval · SHREC17 | retrieval performance | No clear effect |
| Represent the input point cloud and output surface with NUFT, then train a U-Net for reconstruction | Point-to-surface reconstruction · ShapeNet55 Core: car, sofa, bottle | Chamfer distance | Helped |
| Use NUFT point input with noisy point clouds | Point-to-surface reconstruction · ShapeNet55 Core: car, sofa, bottle | distance metrics | Helped |
Closest papers by meaning
- Geometric Deep Learning on Graphs and Manifolds Using Mixture Model CNNs (CVPR 2017 · similarity 0.89)
- Shape correspondence using anisotropic Chebyshev spectral CNNs (CVPR 2020 · similarity 0.87)
- TextureNet: Consistent Local Parametrizations for Learning From High-Resolution Signals on Meshes (CVPR 2019 · similarity 0.87)
- Consistent Geometric Deep Learning via Hilbert Bundles and Cellular Sheaves (NeurIPS 2026 · similarity 0.86)
- HoloNets: Spectral Convolutions do extend to Directed Graphs (ICLR 2024 · similarity 0.86)
- Building Transformation Layers for Riemannian Neural Networks (NeurIPS 2026 · 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.