Research map / Adversarial robustness and security / Attacks on text and graphs
Adversarially-Aware Robust Object Detector
Ziyi Dong, Pengxu Wei, Liang Lin
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 |
|---|---|---|---|
| add AID triplet loss L_aid (ablation) | RobustDet · PASCAL VOC 2007 test | mAP under A_cls: 37.3 -> 41.5 | Improved |
| add AID triplet loss L_aid (ablation) | RobustDet · PASCAL VOC 2007 test | mAP under CWA: 37.9 -> 42.4 | Improved |
| add AID triplet loss L_aid (ablation) | RobustDet · PASCAL VOC 2007 test | mAP clean: 74.9 -> 75.4 | Improved |
| add reconstruction loss L_re to CFR (ablation) | RobustDet* · PASCAL VOC 2007 test | mAP under A_cls: 27.5 -> 45.9 | Improved |
| add reconstruction loss L_re to CFR (ablation) | RobustDet* · PASCAL VOC 2007 test | mAP under CWA: 28.6 -> 48.0 | Improved |
| add CFR to RobustDet | RobustDet · PASCAL VOC 2007 test | mAP under A_cls: 41.5 -> 45.9 | Improved |
| add CFR to RobustDet | RobustDet · PASCAL VOC 2007 test | mAP clean: 75.4 -> 74.8 | No clear effect |
| add CFR to RobustDet | RobustDet · MS-COCO 2017 test | mAP under A_cls: 20.6 -> 20.0 | Hurt |
| add CFR to RobustDet | RobustDet · MS-COCO 2017 test | mAP under DAG: 24.5 -> 16.5 | Hurt |
| use RobustDet instead of adversarially trained MTD | SSD · PASCAL VOC 2007 test | mAP clean: 48.0 -> 75.4 | Improved |
| use RobustDet instead of MTD | SSD · PASCAL VOC 2007 test | mAP under A_cls: 29.1 -> 41.5 | Improved |
| use RobustDet instead of CWAT | SSD · MS-COCO 2017 test | mAP clean: 23.7 -> 36.7 | Improved |
Closest papers by meaning
- Towards Adversarially Robust Object Detection (ICCV 2019 · similarity 0.92)
- Class-Aware Robust Adversarial Training for Object Detection (CVPR 2021 · similarity 0.90)
- Robust and Accurate Object Detection via Adversarial Learning (CVPR 2021 · similarity 0.90)
- KDAT: Inherent Adversarial Robustness via Knowledge Distillation with Adversarial Tuning for Object Detection Models (AAAI 2025 · similarity 0.88)
- Transferable Adversarial Attacks for Object Detection Using Object-Aware Significant Feature Distortion (AAAI 2024 · similarity 0.88)
- Adversarial Training Can Provably Improve Robustness: Theoretical Analysis of Feature Learning Process Under Structured Data (ICLR 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.
