Research map / Graph learning / GNN architectures
SelectionConv: Convolutional Neural Networks for Non-Rectilinear Image Data
David Hart, Michael Whitney, Bryan Morse
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 SelectionConv on a spherical graph instead of applying the FCN to equirectangular images | FCN ResNet-50 · Stanford 2D-3D-S validation | average IoU: 32.57 -> 36.29 | Improved |
| use SelectionConv instead of distortion-aware convolutions | FCN ResNet-50 · Stanford 2D-3D-S validation | average IoU: 34.56 -> 36.29 | Improved |
| copy CNN weights into SelectionConv graph network | VGG-11 · CIFAR-10 validation | accuracy: 84.5 -> 84.5 | No clear effect |
| add small random spatial perturbations to graph points (ablation) | SelectionConv VGG-11 · CIFAR-10 validation | accuracy | No clear effect |
| use SelectionConv spherical graph instead of optimization-based stylization | feed-forward style transfer · 360-degree style transfer | runtime: 8-10 minutes -> 15-20 seconds | Improved |
| process SLIC superpixel graph instead of full-resolution image on CPU | monocular depth estimator · 4K depth prediction | runtime: 25.9 -> 5.1 seconds | Improved |
Closest papers by meaning
- Learning Convolutional Neural Networks for Graphs (ICML 2016 · similarity 0.86)
- Geometrically Principled Connections in Graph Neural Networks (CVPR 2020 · similarity 0.86)
- Geometric Deep Learning on Graphs and Manifolds Using Mixture Model CNNs (CVPR 2017 · similarity 0.86)
- SplineCNN: Fast Geometric Deep Learning With Continuous B-Spline Kernels (CVPR 2018 · similarity 0.86)
- pathGCN: Learning General Graph Spatial Operators from Paths (ICML 2022 · similarity 0.85)
- FeaStNet: Feature-Steered Graph Convolutions for 3D Shape Analysis (CVPR 2018 · 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.
