Research map / Adversarial robustness and security / Attacks on text and graphs
FCA: Learning a 3D Full-Coverage Vehicle Camouflage for Multi-View Physical Adversarial Attack
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 |
|---|---|---|---|
| Apply full-coverage 3D camouflage optimized with FCA instead of the strongest listed baseline | Vehicle object detection · Digital attack, YOLO-V5 | P@0.5 (%): 60.52 -> 32.07 | Helped |
| Apply full-coverage 3D camouflage optimized with FCA instead of the strongest listed baseline | Vehicle object detection · Digital attack, Faster R-CNN | P@0.5 (%): 51.43 -> 34.00 | Helped |
| Apply full-coverage 3D camouflage optimized with FCA instead of the strongest listed baseline | Vehicle object detection · Digital attack, SSD | P@0.5 (%): 49.93 -> 28.67 | Helped |
| Apply full-coverage 3D camouflage optimized with FCA instead of the strongest listed baseline | Vehicle object detection · Digital attack, Mask R-CNN | P@0.5 (%): 52.07 -> 30.80 | Helped |
| Apply full-coverage 3D camouflage optimized with FCA instead of the strongest listed baseline | Vehicle object detection · Physical attack, YOLO-V5 | P@0.5 (%): 72.22 -> 65.28 | Helped |
| Apply full-coverage 3D camouflage optimized with FCA instead of the strongest listed baseline | Vehicle object detection · Physical attack, Faster R-CNN | P@0.5 (%): 43.06 -> 24.31 | Helped |
| Apply full-coverage 3D camouflage optimized with FCA instead of the strongest listed baseline | Vehicle object detection · Physical attack, SSD | P@0.5 (%): 40.97 -> 29.17 | Helped |
| Apply full-coverage 3D camouflage optimized with FCA instead of the strongest listed baseline | Vehicle object detection · Physical attack, Mask R-CNN | P@0.5 (%): 45.83 -> 29.17 | Helped |
| Use IoU adversarial loss alone, retaining smooth loss (ablation) | FCA · YOLO-V5 loss ablation | ASR (%): 87.67 -> 90.96 | Helped |
| Use classification adversarial loss alone, retaining smooth loss (ablation) | FCA · YOLO-V5 loss ablation | ASR (%): 87.67 -> 82.83 | Hurt |
| Use objectness adversarial loss alone, retaining smooth loss (ablation) | FCA · YOLO-V5 loss ablation | ASR (%): 87.67 -> 46.16 | Hurt |
| Combine classification and objectness adversarial losses, retaining smooth loss (ablation) | FCA · YOLO-V5 loss ablation | ASR (%): 87.67 -> 84.49 | Hurt |
Closest papers by meaning
- CNCA: Toward Customizable and Natural Generation of Adversarial Camouflage for Vehicle Detectors (NeurIPS 2024 · similarity 0.92)
- RAUCA: A Novel Physical Adversarial Attack on Vehicle Detectors via Robust and Accurate Camouflage Generation (ICML 2024 · similarity 0.90)
- PhyCamo: A Robust Physical Camouflage via Contrastive Learning for Multi-View Physical Adversarial Attack (AAAI 2025 · similarity 0.87)
- Infrared Adversarial Car Stickers (CVPR 2024 · similarity 0.86)
- Physical 3D Adversarial Attacks against Monocular Depth Estimation in Autonomous Driving (CVPR 2024 · similarity 0.84)
- Adversarial Attacks on Event-Based Pedestrian Detectors: A Physical Approach (AAAI 2025 · 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.