Research map / Point clouds / Point cloud classification
Robust Structured Declarative Classifiers for 3D Point Clouds: Defending Adversarial Attacks With Implicit Gradients
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 LPC with a structured declarative lattice defender instead of PointNet | Point cloud classification · ModelNet40, JGBA | test accuracy (%): 5.37 -> 89.51 | Helped |
| Use LPC with a structured declarative lattice defender instead of PointNet | Point cloud classification · ScanNet, JGBA | test accuracy (%): 13.45 -> 83.16 | Helped |
| Use LPC instead of PointNet | Point cloud classification · ModelNet40, FGSM | attack success rate (%): 48.99 -> 0.00 | Helped |
| Use LPC instead of PointNet | Point cloud classification · ScanNet, FGSM | attack success rate (%): 46.03 -> 0.00 | Helped |
| Use LPC instead of PointNet | Point cloud classification · ModelNet40, LG-GAN | attack success rate (%): 97.0 -> 0.8 | Helped |
| Add T-Net (ablation) | LPC with ResNet50 · ModelNet40 | test accuracy (%): 88.9 -> 87.3 | Hurt |
| Add T-Net (ablation) | LPC with EfficientNet-B5 · ModelNet40 | test accuracy (%) | Hurt |
| Binarize barycentric weights (ablation) | LPC with ResNet50 · ModelNet40 | test accuracy (%): 88.2 -> 88.9 | Helped |
| Binarize barycentric weights (ablation) | LPC with EfficientNet-B5 · ModelNet40 | test accuracy (%) | Helped |
| Binarize barycentric weights (ablation) | LPC with EfficientNet-B5 · ScanNet | test accuracy (%): 80.5 -> 80.0 | Hurt |
| Apply random rotation augmentation (ablation) | LPC with ResNet50 · ModelNet40 | test accuracy (%): 88.2 -> 60.0 | Hurt |
| Apply random rotation augmentation (ablation) | LPC with EfficientNet-B5 · ModelNet40 | test accuracy (%): 89.5 -> 83.3 | Hurt |
Closest papers by meaning
- PointCVaR: Risk-Optimized Outlier Removal for Robust 3D Point Cloud Classification (AAAI 2024 · similarity 0.89)
- CausalPC: Improving the Robustness of Point Cloud Classification by Causal Effect Identification (CVPR 2024 · similarity 0.88)
- Curvature-Invariant Adversarial Attacks for 3D Point Clouds (AAAI 2024 · similarity 0.87)
- Explicitly Perceiving and Preserving the Local Geometric Structures for 3D Point Cloud Attack (AAAI 2024 · similarity 0.87)
- Improving Robustness of 3D Point Cloud Recognition from a Fourier Perspective (NeurIPS 2024 · similarity 0.87)
- Hide in Thicket: Generating Imperceptible and Rational Adversarial Perturbations on 3D Point Clouds (CVPR 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.