Research map / Point clouds / Shape and surface completion
GeoNet: Deep Geodesic Networks for Point Cloud Analysis
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 |
|---|---|---|---|
| Add GeoNet geodesic features to PU-Net (PUF) | PU-Net · ShapeNet heldout training categories, upsampling | MSE (x10000): 7.14 -> 6.23 | Helped |
| Add GeoNet geodesic features to PU-Net (PUF) | PU-Net · ShapeNet leftout categories, upsampling | EMD: 11.43 -> 8.90 | Helped |
| Add geodesic-aware features through PointNet++ fusion (POF) | PointNet++ · Shrec15, normal estimation | accuracy at ≤ 15° (%): 75.49±1.95 -> 79.66±1.21 | Helped |
| Add geodesic-aware features through PointNet++ fusion (POF) | PointNet++ · ShapeNet leftout categories, normal estimation | accuracy at ≤ 15° (%): 51.17 -> 53.78 | Helped |
| Add geodesic-aware features through PointNet++ fusion (POF) | PointNet++ · Shrec15, non-rigid shape classification | accuracy (%): 73.56 -> 94.67 | Helped |
| Add geodesic-aware features through PointNet++ fusion (POF) | PointNet++ · Noisy Shrec15 point clouds, Gaussian noise 1.2% | accuracy (%): 62.38 -> 84.98 | Helped |
| Estimate geodesic neighborhoods with GeoNet rather than KNN-Graph K-3 | geodesic neighborhood estimation · ShapeNet heldout training categories, v1, r ≤ 0.1 | MSE (x100): 8.75 -> 5.67 | Helped |
Closest papers by meaning
- Modeling Local Geometric Structure of 3D Point Clouds Using Geo-CNN (CVPR 2019 · similarity 0.90)
- NeuroGF: A Neural Representation for Fast Geodesic Distance and Path Queries (NeurIPS 2023 · similarity 0.88)
- From Extrinsic to Intrinsic: Geodesic-Guided Representation Learning for 3D Geometric Data (ICML 2026 · similarity 0.88)
- Deep Geometric Prior for Surface Reconstruction (CVPR 2019 · similarity 0.88)
- TearingNet: Point Cloud Autoencoder To Learn Topology-Friendly Representations (CVPR 2021 · similarity 0.87)
- Geodesic Self-Attention for 3D Point Clouds (NeurIPS 2022 · similarity 0.87)
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.