Research map / Point clouds / Shape and surface completion
Unsupervised Learning of Intrinsic Structural Representation Points
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 |
|---|---|---|---|
| Use unsupervised structure-point label transfer instead of BAE-Net few-shot transfer | ShapeNet part label transfer with 3 exemplars · ShapeNet part categories | average IoU (%) | Mixed |
| Train on 2048 input points and test with different uniformly sampled input densities | Proposed structure-point network · ShapeNet Chair, Airplane, Car, and Motorbike | point-wise average Euclidean distance (%): 0.7346, 0.1308, 0.5470, 0.3070 (256); 0.1308, 0.0169, 0.1348, 0.0610 (512); 0.0200, 0.0022, 0.0099, 0.0195 (1024); 0.002 | No clear effect |
| Test on non-uniformly sampled inputs with holes | Proposed structure-point network · ShapeNet categories used in the sampling robustness experiment | point-wise average Euclidean distance (%): 0.3499% | No clear effect |
| Replace the PointNet++ encoder with PointConv (ablation) | Proposed point integration module · 3D semantic correspondence | correspondence accuracy | No clear effect |
Closest papers by meaning
- StructEdit: Learning Structural Shape Variations (CVPR 2020 · similarity 0.89)
- Shape Unicode: A Unified Shape Representation (CVPR 2019 · similarity 0.89)
- Unsupervised 3D Structure Inference from Category-Specific Image Collections (CVPR 2024 · similarity 0.88)
- Learning Partonomic 3D Reconstruction from Image Collections (CVPR 2025 · similarity 0.87)
- X-3D: Explicit 3D Structure Modeling for Point Cloud Recognition (CVPR 2024 · similarity 0.87)
- PolyhedronNet: Representation Learning for Polyhedra with Surface-attributed Graph (ICLR 2025 · 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.