Research map / Point clouds / Point cloud classification
Geometry Sharing Network for 3D Point Cloud Classification and Segmentation
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 |
|---|---|---|---|
| Combine Euclidean-space and eigenvalue-space neighbor grouping (ablation) | GS-Net · ModelNet40 | accuracy: 92.5 -> 92.8 | Helped |
| Use FPS down-sampling (ablation) | GS-Net · ModelNet40, 1024 points | accuracy: 92.5 -> 92.8 | Helped |
| Increase point count from 1024 to 2048 (ablation) | GS-Net · ModelNet40 | accuracy: 92.8 -> 92.9 | Helped |
| Add coordinate differences and coordinates to coordinates alone (ablation) | GS-Net · ModelNet40 | accuracy: 92.5 -> 92.7 | Helped |
| Add eigenvalue differences and eigenvalues to coordinate features (ablation) | GS-Net · ModelNet40 | accuracy: 92.7 -> 92.8 | Helped |
| Add Euclidean distance to coordinate and eigenvalue features (ablation) | GS-Net · ModelNet40 | accuracy: 92.8 -> 92.9 | Helped |
| Add eigenvector differences and eigenvectors (ablation) | GS-Net · ModelNet40 | accuracy: 92.9 -> 92.4 | Hurt |
| Use eigenvalue features only, without coordinates (ablation) | GS-Net · ModelNet40 z/s | accuracy: 37.1 -> 72.8 | Helped |
| Use eigenvalue features only, without coordinates (ablation) | GS-Net · ModelNet40 0/s | accuracy: 19.4 -> 50.7 | Helped |
| Use eigenvalue features only, without coordinates (ablation) | GS-Net · ModelNet40 s/s | accuracy: 87.9 -> 82.8 | Hurt |
Closest papers by meaning
- Convolution in the Cloud: Learning Deformable Kernels in 3D Graph Convolution Networks for Point Cloud Analysis (CVPR 2020 · similarity 0.88)
- Learning to Segment 3D Point Clouds in 2D Image Space (CVPR 2020 · similarity 0.87)
- Modeling Local Geometric Structure of 3D Point Clouds Using Geo-CNN (CVPR 2019 · similarity 0.86)
- GeoNet: Deep Geodesic Networks for Point Cloud Analysis (CVPR 2019 · similarity 0.86)
- DGC-GNN: Leveraging Geometry and Color Cues for Visual Descriptor-Free 2D-3D Matching (CVPR 2024 · similarity 0.85)
- LinNet: Linear Network for Efficient Point Cloud Representation Learning (NeurIPS 2024 · 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.