Research map / Point clouds / Point cloud classification
Symmetria: A Synthetic Dataset for Learning in Point Clouds
Ivan Sipiran, Gustavo Santelices, Lucas Oyarzún, Andrea Ranieri, Chiara Romanengo, Silvia Biasotti, B. Falcidieno
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 |
|---|---|---|---|
| pre-train on SymSSL-10K instead of ShapeNet | Point-MAE · ModelNet40 k-NN | accuracy: 78.4 -> 81.7 | Improved |
| pre-train on SymSSL-10K instead of ShapeNet | Point-MAE · ModelNet40 fine-tuning | accuracy: 93.8 -> 92.5 | Hurt |
| pre-train on SymSSL-10K instead of ShapeNet | PointGPT-S · ModelNet40 k-NN | accuracy: 81.6 -> 66.5 | Hurt |
| pre-train on SymSSL-10K instead of ShapeNet | PointGPT-S · ScanObjectNN OBJ-ONLY | accuracy: 90.0 -> 79.0 | Hurt |
| pre-train on SymSSL-50K instead of ShapeNet | PointGPT-S · ScanObjectNN PB-T50-RS | accuracy: 86.9 -> 84.9 | Mixed |
| pre-train on SymSSL-10K instead of ShapeNet | Point-MAE · ShapeNetPart instance mIoU | mIoU: 86.1 -> 85.9 | No clear effect |
| increase SymSSL size from 10K to 50K | Point-MAE · ScanObjectNN OBJ-ONLY | accuracy: 87.8 -> 86.6 | Hurt |
| increase SymSSL size from 10K to 1M | Point-MAE · ModelNet40 k-NN | accuracy: 81.7 -> 73.3 | Hurt |
| increase SymSSL size from 10K to 1M | Point-MAE · ModelNet40 fine-tuning | accuracy: 92.5 -> 92.8 | No clear effect |
| add geometric perturbations (vs clean only) to SymSSL-10K (ablation) | Point-MAE · ModelNet40 k-NN | accuracy: 78.6 -> 81.7 | Improved |
| add complex curve families (vs simple subset) to SymSSL-10K (ablation) | Point-MAE · ModelNet40 k-NN | accuracy: 78.6 -> 79.6 | Improved |
| add geometric perturbations (vs clean only) to SymSSL-10K (ablation) | Point-MAE · ShapeNetPart class mIoU | mIoU: 83.7 -> 84.2 | Improved |
Closest papers by meaning
- SymmCompletion: High-Fidelity and High-Consistency Point Cloud Completion with Symmetry Guidance (AAAI 2025 · similarity 0.89)
- Point-Cache: Test-time Dynamic and Hierarchical Cache for Robust and Generalizable Point Cloud Analysis (CVPR 2025 · similarity 0.88)
- Rethinking Point Cloud Data Augmentation: Topologically Consistent Deformation (ICML 2025 · similarity 0.88)
- SMART-PC: Skeletal Model Adaptation for Robust Test-Time Training in Point Clouds (ICML 2025 · similarity 0.87)
- Point Cloud Dataset Distillation (ICML 2025 · similarity 0.87)
- A Linear Time and Space Local Point Cloud Geometry Encoder via Vectorized Kernel Mixture (VecKM) (ICML 2024 · 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.
