Research map / Adversarial robustness and security / Certified and empirical robustness
Robust Tracking against Adversarial Attacks
Shuai Jia, Chao Ma, Yibing Song, Xiaokang Yang
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 |
|---|---|---|---|
| apply adversarial attack | DaSiamRPN · OTB100 | AUC: 0.655 -> 0.050 | Hurt |
| apply adversarial attack | RT-MDNet · OTB100 | AUC: 0.643 -> 0.131 | Hurt |
| add adversarial defense after attack | DaSiamRPN · OTB100 | AUC: 0.050 -> 0.721 | Improved |
| add adversarial defense after attack | RT-MDNet · OTB100 | AUC: 0.131 -> 0.653 | Improved |
| apply defense on clean sequences | DaSiamRPN · OTB100 | AUC: 0.880 -> 0.886 | Improved |
| apply defense on clean sequences | RT-MDNet · OTB100 | AUC: 0.876 -> 0.883 | Improved |
| apply adversarial attack | DaSiamRPN · VOT-2018 | EAO: 0.380 -> 0.097 | Hurt |
| add defense after attack | DaSiamRPN · VOT-2018 | EAO: 0.097 -> 0.195 | Improved |
| apply adversarial attack | RT-MDNet · VOT-2018 | EAO: 0.176 -> 0.076 | Hurt |
| add defense after attack | RT-MDNet · VOT-2018 | EAO: 0.076 -> 0.110 | Improved |
| use adversarial instead of random perturbation | DaSiamRPN · VOT-2018 | EAO: 0.223 -> 0.097 | Hurt |
| add defense to clean tracker | DaSiamRPN · VOT-2016 | EAO: 0.439 -> 0.418 | No clear effect |
Closest papers by meaning
- One-Shot Adversarial Attacks on Visual Tracking With Dual Attention (CVPR 2020 · similarity 0.92)
- IoU Attack: Towards Temporally Coherent Black-Box Adversarial Attack for Visual Object Tracking (CVPR 2021 · similarity 0.91)
- Enhancing Tracking Robustness with Auxiliary Adversarial Defense Networks (ECCV 2024 · similarity 0.90)
- Efficient Adversarial Attacks for Visual Object Tracking (ECCV 2020 · similarity 0.89)
- Learning To Adversarially Blur Visual Object Tracking (ICCV 2021 · similarity 0.88)
- SPARK: Spatial-aware Online Incremental Attack Against Visual Tracking (ECCV 2020 · similarity 0.88)
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.
