Research map / Point clouds / Point cloud classification
Regularization Strategy for Point Cloud via Rigidly Mixed Sample
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 RSMix | PointNet++ · ModelNet40 | accuracy: 91.0 -> 91.6 | Helped |
| Add RSMix | DGCNN · ModelNet10 | accuracy: 94.8 -> 95.9 | Helped |
| Use shape-preserving RSMix rather than PointMixup | DGCNN · ModelNet40 | accuracy: 93.1 -> 93.5 | Helped |
| Use ball-query neighboring rather than KNN (ablation) | DGCNN + RSMix · ModelNet40 | accuracy: 93.3 -> 93.4 | Helped |
| Add RSMix to scaling augmentation (ablation) | DGCNN · ModelNet40 | accuracy: 92.6 -> 93.4 | Helped |
| Add RSMix to scaling augmentation (ablation) | DGCNN · ModelNet10 | accuracy: 94.8 -> 95.4 | Helped |
| Use rotational augmentation during training (ablation) | PointNet++ · ModelNet40 multi-view | accuracy: 78.5 -> 90.8 | Helped |
| Add RSMix (ablation) | PointNet++ without rotational augmentation · ModelNet40 multi-view | accuracy: 78.5 -> 71.5 | Hurt |
| Add RSMix under large jitter noise (σ²=0.05) (ablation) | PointNet++ · ModelNet40 single-view | accuracy: 15.3 -> 13.0 | Hurt |
| Add RSMix under large jitter noise (σ²=0.05) (ablation) | PointNet++ with ConvDA · ModelNet40 single-view | accuracy: 23.4 -> 21.2 | Hurt |
| Add RSMix under scaling noise at scale 0.6 (ablation) | PointNet++ · ModelNet40 multi-view | accuracy: 83.7 -> 82.5 | Hurt |
| Add RSMix under point-drop noise (ablation) | PointNet++ · ModelNet40 | accuracy: 88.2 -> 91.9 | Helped |
Closest papers by meaning
- PointPatchMix: Point Cloud Mixing with Patch Scoring (AAAI 2024 · similarity 0.90)
- Rethinking Point Cloud Data Augmentation: Topologically Consistent Deformation (ICML 2025 · similarity 0.87)
- SPU-IMR: Self-supervised Arbitrary-scale Point Cloud Upsampling via Iterative Mask-recovery Network (AAAI 2025 · similarity 0.84)
- Arbitrary-Scale Point Cloud Upsampling by Voxel-Based Network with Latent Geometric-Consistent Learning (AAAI 2024 · similarity 0.84)
- Learning Continuous Implicit Field with Local Distance Indicator for Arbitrary-Scale Point Cloud Upsampling (AAAI 2024 · similarity 0.84)
- An Explicit Frame Construction for Normalizing 3D Point Clouds (ICML 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.