Research map / Point clouds / Point cloud classification
Self-Supervised Global-Local Structure Modeling for Point Cloud Domain Adaptation With Reliable Voted Pseudo Labels
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 global scale prediction (ablation) | DGCNN without adaptation · ScanNet-10 → ShapeNet-10 | accuracy: 64.2 -> 69.8 | Helped |
| Add local 3D-2D-3D reconstruction (ablation) | DGCNN without adaptation · ScanNet-10 → ShapeNet-10 | accuracy: 64.2 -> 67.9 | Helped |
| Add global scale prediction and local 3D-2D-3D reconstruction (ablation) | DGCNN without adaptation · ScanNet-10 → ShapeNet-10 | accuracy: 64.2 -> 71.2 | Helped |
| Add adversarial training (ablation) | DGCNN without adaptation · ScanNet-10 → ShapeNet-10 | accuracy: 64.2 -> 66.7 | Helped |
| Add global scale prediction and adversarial training (ablation) | DGCNN without adaptation · ScanNet-10 → ShapeNet-10 | accuracy: 64.2 -> 67.4 | Helped |
| Add global scale prediction, local reconstruction, and adversarial training (ablation) | DGCNN without adaptation · ScanNet-10 → ShapeNet-10 | accuracy: 64.2 -> 73.8 | Helped |
| Add reliable voted pseudo-label self-training (ablation) | Scale + 3D-2D-3D + adversarial training · ScanNet-10 → ShapeNet-10 | accuracy: 73.8 -> 76.2 | Helped |
| Use voting without consistency-based reliable-sample selection (ablation) | Scale + 3D-2D-3D + adversarial training · ScanNet-10 → ShapeNet-10 | accuracy: 73.8 -> 66.7 | Hurt |
| Use GLRV instead of the strongest baseline | PointDA · ModelNet-10 → ShapeNet-10 | accuracy: 84.8 ± 0.1 -> 85.4 ± 0.4 | Helped |
| Use GLRV instead of the strongest baseline | PointDA · ModelNet-10 → ScanNet-10 | accuracy: 59.8 ± 0.2 -> 60.4 ± 0.4 | Helped |
| Use GLRV instead of the strongest baseline | PointDA · accuracy | accuracy: 80.8 ± 0.6 -> 78.8 ± 0.6 | Hurt |
| Use GLRV instead of the strongest baseline | PointDA · ShapeNet-10 → ScanNet-10 | accuracy: 56.7 ± 0.2 -> 57.7 ± 0.4 | Helped |
Closest papers by meaning
- PointDAN: A Multi-Scale 3D Domain Adaption Network for Point Cloud Representation (NeurIPS 2019 · similarity 0.90)
- LinNet: Linear Network for Efficient Point Cloud Representation Learning (NeurIPS 2024 · similarity 0.87)
- Point Cloud Pre-training with Diffusion Models (CVPR 2024 · similarity 0.87)
- Construct to Associate: Cooperative Context Learning for Domain Adaptive Point Cloud Segmentation (CVPR 2024 · similarity 0.87)
- PointDGMamba: Domain Generalization of Point Cloud Classification via Generalized State Space Model (AAAI 2025 · similarity 0.87)
- PointLoRA: Low-Rank Adaptation with Token Selection for Point Cloud Learning (CVPR 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.