Enregistré dans:
| Auteurs principaux: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2508.10600 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912537556025344 |
|---|---|
| author | Cao, Yuxin Zhang, Yedi He, Wentao Liao, Yifan Xiao, Yan Li, Chang Huang, Zhiyong Dong, Jin Song |
| author_facet | Cao, Yuxin Zhang, Yedi He, Wentao Liao, Yifan Xiao, Yan Li, Chang Huang, Zhiyong Dong, Jin Song |
| contents | Learning-based autonomous driving systems remain critically vulnerable to adversarial patches, posing serious safety and security risks in their real-world deployment. Black-box attacks, notable for their high attack success rate without model knowledge, are especially concerning, with their transferability extensively studied to reduce computational costs compared to query-based attacks. Previous transferability-based black-box attacks typically adopt mean Average Precision (mAP) as the evaluation metric and design training loss accordingly. However, due to the presence of multiple detected bounding boxes and the relatively lenient Intersection over Union (IoU) thresholds, the attack effectiveness of these approaches is often overestimated, resulting in reduced success rates in practical attacking scenarios. Furthermore, patches trained on low-resolution data often fail to maintain effectiveness on high-resolution images, limiting their transferability to autonomous driving datasets. To fill this gap, we propose P$^3$A, a Powerful and Practical Patch Attack framework for 2D object detection in autonomous driving, specifically optimized for high-resolution datasets. First, we introduce a novel metric, Practical Attack Success Rate (PASR), to more accurately quantify attack effectiveness with greater relevance for pedestrian safety. Second, we present a tailored Localization-Confidence Suppression Loss (LCSL) to improve attack transferability under PASR. Finally, to maintain the transferability for high-resolution datasets, we further incorporate the Probabilistic Scale-Preserving Padding (PSPP) into the patch attack pipeline as a data preprocessing step. Extensive experiments show that P$^3$A outperforms state-of-the-art attacks on unseen models and unseen high-resolution datasets, both under the proposed practical IoU-based evaluation metric and the previous mAP-based metrics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_10600 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Towards Powerful and Practical Patch Attacks for 2D Object Detection in Autonomous Driving Cao, Yuxin Zhang, Yedi He, Wentao Liao, Yifan Xiao, Yan Li, Chang Huang, Zhiyong Dong, Jin Song Computer Vision and Pattern Recognition Learning-based autonomous driving systems remain critically vulnerable to adversarial patches, posing serious safety and security risks in their real-world deployment. Black-box attacks, notable for their high attack success rate without model knowledge, are especially concerning, with their transferability extensively studied to reduce computational costs compared to query-based attacks. Previous transferability-based black-box attacks typically adopt mean Average Precision (mAP) as the evaluation metric and design training loss accordingly. However, due to the presence of multiple detected bounding boxes and the relatively lenient Intersection over Union (IoU) thresholds, the attack effectiveness of these approaches is often overestimated, resulting in reduced success rates in practical attacking scenarios. Furthermore, patches trained on low-resolution data often fail to maintain effectiveness on high-resolution images, limiting their transferability to autonomous driving datasets. To fill this gap, we propose P$^3$A, a Powerful and Practical Patch Attack framework for 2D object detection in autonomous driving, specifically optimized for high-resolution datasets. First, we introduce a novel metric, Practical Attack Success Rate (PASR), to more accurately quantify attack effectiveness with greater relevance for pedestrian safety. Second, we present a tailored Localization-Confidence Suppression Loss (LCSL) to improve attack transferability under PASR. Finally, to maintain the transferability for high-resolution datasets, we further incorporate the Probabilistic Scale-Preserving Padding (PSPP) into the patch attack pipeline as a data preprocessing step. Extensive experiments show that P$^3$A outperforms state-of-the-art attacks on unseen models and unseen high-resolution datasets, both under the proposed practical IoU-based evaluation metric and the previous mAP-based metrics. |
| title | Towards Powerful and Practical Patch Attacks for 2D Object Detection in Autonomous Driving |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2508.10600 |