Research map / Adversarial robustness and security / Attacks on text and graphs
CAP: Robust Point Cloud Classification via Semantic and Structural Modeling
Read the paperCheck 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 CAP | PointNet · ModelNet40, 9 attacks (average) | attack success rate: 70.2% -> 2.7% | Helped |
| Add CAP | DGCNN · ModelNet40, 9 attacks (average) | attack success rate: 65.4% -> 3.5% | Helped |
| Add CAP | PointCNN · ModelNet40, 9 attacks (average) | attack success rate: 16.7% -> 2.0% | Helped |
| Add CAP | PointNet · ModelNet40 | mean test accuracy: 86.2% -> 86.8% | Helped |
| Add CAP | DGCNN · ModelNet40 | mean test accuracy: 88.9% -> 90.1% | Helped |
| Add CAP | PointCNN · ModelNet40 | mean test accuracy: 86.8% -> 87.9% | Helped |
| Add CAP | PointNet · ShapeNet | mean test accuracy: 78.6% -> 77.1% | Hurt |
| Add CAP | DGCNN · ShapeNet | mean test accuracy: 80.5% -> 80.5% | No clear effect |
| Add CAP | PointCNN · ShapeNet | mean test accuracy: 77.9% -> 79.4% | Helped |
| Add CAP | PointNet · ShapeNet, 6 attacks (average) | attack success rate: 54.4% -> 1.5% | Helped |
| Add CAP | DGCNN · ShapeNet, 6 attacks (average) | attack success rate: 49.2% -> 3.4% | Helped |
| Add CAP | PointCNN · ShapeNet, 6 attacks (average) | attack success rate: 7.2% -> 1.8% | Helped |
Closest papers by meaning
- Robust Structured Declarative Classifiers for 3D Point Clouds: Defending Adversarial Attacks With Implicit Gradients (CVPR 2022 · similarity 0.91)
- A Critical Revisit of Adversarial Robustness in 3D Point Cloud Recognition with Diffusion-Driven Purification (ICML 2023 · similarity 0.89)
- CausalPC: Improving the Robustness of Point Cloud Classification by Causal Effect Identification (CVPR 2024 · similarity 0.89)
- Improving Robustness of 3D Point Cloud Recognition from a Fourier Perspective (NeurIPS 2024 · similarity 0.87)
- PointCVaR: Risk-Optimized Outlier Removal for Robust 3D Point Cloud Classification (AAAI 2024 · similarity 0.86)
- Hide in Thicket: Generating Imperceptible and Rational Adversarial Perturbations on 3D Point Clouds (CVPR 2024 · similarity 0.86)
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.