Research map / Adversarial robustness and security / Attacks on text and graphs
One-Shot Adversarial Attacks on Visual Tracking With Dual Attention
Xuesong Chen, Xiyu Yan, Feng Zheng, Yong Jiang, Shu-Tao Xia, Yong Zhao, Rongrong Ji
Check 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 the one-shot dual-attention perturbation to the initial target patch | Siamese trackers · OTB100 | precision: 91.4 -> 33.7 | Hurt |
| Apply the one-shot dual-attention perturbation to the initial target patch | Siamese trackers · LaSOT | precision: 50.2 -> 12.2 | Hurt |
| Apply the one-shot dual-attention perturbation to the initial target patch | Siamese trackers · GOT10K | AO: 65.1 -> 31.2 | Hurt |
| Add confidence attention to batch confidence loss (ablation) | One-shot attack on SiamRPN++(R) · OTB100 | precision: 38.8 -> 37.1 | Improved |
| Add feature attention to feature loss (ablation) | One-shot attack on SiamRPN++(R) · OTB100 | precision: 38.7 -> 34.3 | Improved |
| Combine batch confidence and feature losses without attention (ablation) | One-shot attack on SiamRPN++(R) · OTB100 | precision: 38.8 -> 37.5 | Improved |
| Use both attention-weighted losses together (ablation) | One-shot attack on SiamRPN++(R) · OTB100 | precision: 37.1 -> 33.7 | Improved |
| Use dual attention instead of combining the two losses without attention (ablation) | One-shot attack on SiamRPN++(R) · OTB100 | precision: 37.5 -> 33.7 | Improved |
Closest papers by meaning
- Robust Tracking against Adversarial Attacks (ECCV 2020 · similarity 0.92)
- Efficient Adversarial Attacks for Visual Object Tracking (ECCV 2020 · similarity 0.91)
- IoU Attack: Towards Temporally Coherent Black-Box Adversarial Attack for Visual Object Tracking (CVPR 2021 · similarity 0.91)
- Towards Universal Physical Attacks on Single Object Tracking (AAAI 2021 · similarity 0.90)
- SPARK: Spatial-aware Online Incremental Attack Against Visual Tracking (ECCV 2020 · similarity 0.89)
- A Unified Multi-Scenario Attacking Network for Visual Object Tracking (AAAI 2021 · similarity 0.89)
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.
