Research map / Adversarial robustness and security / Attacks on text and graphs
Physical-World Optical Adversarial Attacks on 3D Face Recognition
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 direction constraint, renormalization, and 3D transform-invariant loss (ablation) | Phase Shifting Attack · Targeted attacks across five 3D classifiers | ASR: +34% on Pointnet, 20% on Pointnet++ SSG, 24% on Pointnet++ MSG, 20% on DGCNN, and 33% on CurveNet | Helped |
| Add direction constraint, renormalization, and 3D transform-invariant loss (ablation) | Phase Shifting Attack · Untargeted attacks | ASR: 27% | Helped |
| Add 3D transform-invariant loss (ablation) | Phase Shifting Attack · Rotated point clouds | target-label prediction robustness | Helped |
| Add 3D transform-invariant loss (ablation) | Phase Shifting Attack · Dodging attack transferability | transferability | Helped |
| Add 3D transform-invariant loss (ablation) | Phase Shifting Attack · Perturbation size | perturbation size | Hurt |
Closest papers by meaning
- ProjAttacker: A Configurable Physical Adversarial Attack for Face Recognition via Projector (CVPR 2025 · similarity 0.89)
- Hide in Thicket: Generating Imperceptible and Rational Adversarial Perturbations on 3D Point Clouds (CVPR 2024 · similarity 0.86)
- Towards Transferable Targeted 3D Adversarial Attack in the Physical World (CVPR 2024 · similarity 0.85)
- Omni-Angle Assault: An Invisible and Powerful Physical Adversarial Attack on Face Recognition (ICML 2025 · similarity 0.85)
- Seeing is Not Believing: Adversarial Natural Object Optimization for Hard-Label 3D Scene Attacks (CVPR 2025 · similarity 0.84)
- Curvature-Invariant Adversarial Attacks for 3D Point Clouds (AAAI 2024 · similarity 0.84)
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.