Research map / Point clouds / Point cloud classification
PVRNet: Point-View Relation Neural Network for 3D Shape Recognition
Haoxuan You, Yifan Feng, Xibin Zhao, Changqing Zou, Rongrong Ji, Yue Gao
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 |
|---|---|---|---|
| Fuse point-cloud features with views using learned point-view relations | PVRNet · ModelNet40 | overall accuracy: 92.62\% -> 93.61\% | Improved |
| Fuse point-cloud features with views using learned point-view relations | PVRNet · ModelNet40 | mean class accuracy: 90.81\% -> 91.64\% | Improved |
| Add point-single-view fusion to the point-cloud and multi-view feature extractors (ablation) | PVRNet · ModelNet40 | overall accuracy: 92.62\% -> 92.84\% | Improved |
| Add point-multi-view fusion to point-single-view fusion, with a 2-view set (ablation) | PVRNet · ModelNet40 | overall accuracy: 92.84\% -> 92.99\% | Improved |
| Increase the point-multi-view fusion sets from up to 2 views to up to 3 views (ablation) | PVRNet · ModelNet40 | overall accuracy: 92.99\% -> 93.47\% | Improved |
| Increase the point-multi-view fusion sets from up to 3 views to up to 4 views (ablation) | PVRNet · ModelNet40 | overall accuracy: 93.47\% -> 93.61\% | Improved |
| Reduce the test input from 12 views to 4 views (ablation) | PVRNet · ModelNet40 | overall accuracy: 93.6\% -> 91.5\% | Hurt |
Closest papers by meaning
- Learning Relationships for Multi-View 3D Object Recognition (ICCV 2019 · similarity 0.86)
- Relation-Shape Convolutional Neural Network for Point Cloud Analysis (CVPR 2019 · similarity 0.85)
- ViewNet: A Novel Projection-Based Backbone With View Pooling for Few-Shot Point Cloud Classification (CVPR 2023 · similarity 0.85)
- MVTN: Multi-View Transformation Network for 3D Shape Recognition (ICCV 2021 · similarity 0.85)
- Multi-View Representation is What You Need for Point-Cloud Pre-Training (ICLR 2024 · similarity 0.84)
- PointAttN: You Only Need Attention for Point Cloud Completion (AAAI 2024 · similarity 0.83)
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.
