Research map / Adversarial robustness and security / Attacks on text and graphs
Physical Adversarial Attack Meets Computer Vision: A Decade Survey
Hui Wei, Hao Tang, Xuemei Jia, Zhixiang Wang, Hanxun Yu, Zhubo Li, Shin’ichi Satoh, Luc Van Gool, Zheng Wang
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 |
|---|---|---|---|
| upgrade AdvEyeglass to AdvEyeglass+ | AdvEyeglass · Face recognition (physical attack) | hiPAA: 0.75 -> 0.83 | Improved |
| upgrade AdvACO to AdvACO+ | AdvACO · General classification (physical attack) | hiPAA: 0.71 -> 0.81 | Improved |
| add cross-model, cross-scenario and all transformation factors (TC-EGA vs AdvYOLO) | person detection attacks · Person detection (physical attack) | Robustness: 0.50 -> 1.00 | Improved |
| prioritize effectiveness over robustness (CPAttack) | CPAttack · General classification (physical attack) | Robustness / Effectiveness: Eff. 1.00, Rob. 0.17, hiPAA 0.57 | Mixed |
| trade economics for effectiveness (AdvISP) | AdvISP · General classification (physical attack) | hiPAA: Eff. 0.90, Eco. 0.00, hiPAA 0.49 | Hurt |
| newer attacks versus older ones on classification | classification attacks · General classification (physical attack) | hiPAA: EOT 0.80 vs ViewFool 0.81 | No clear effect |
| use large-area light projection (SLAP) rather than patches | SLAP · Traffic sign detection (physical attack) | Effectiveness / Stealthiness: Eff. 0.99, Ste. 0.25, hiPAA 0.73 | Mixed |
| extend RP2 from classification to detection on YOLOv2 | RP2 · Traffic sign detection (physical attack) | ASR: 85.6% ASR indoor, 72.5% outdoor | Improved |
Closest papers by meaning
- Connecting the Digital and Physical World: Improving the Robustness of Adversarial Attacks (AAAI 2019 · similarity 0.91)
- Robust Physical-World Attacks on Deep Learning Visual Classification (CVPR 2018 · similarity 0.90)
- Beyond Digital Domain: Fooling Deep Learning Based Recognition System in Physical World (AAAI 2020 · similarity 0.90)
- Revisiting Adversarial Patches for Designing Camera-Agnostic Attacks against Person Detection (NeurIPS 2024 · similarity 0.88)
- Comparing the Robustness of Modern No-Reference Image- and Video-Quality Metrics to Adversarial Attacks (AAAI 2024 · similarity 0.88)
- Breaking Barriers in Physical-World Adversarial Examples: Improving Robustness and Transferability via Robust Feature (AAAI 2025 · 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.
