Fool the Stoplight: Realistic Adversarial Patch Attacks on Traffic Light Detectors
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866910989370261504 |
|---|---|
| author | Pavlitska, Svetlana Robb, Jamie Polley, Nikolai Yazgan, Melih Zöllner, J. Marius |
| author_facet | Pavlitska, Svetlana Robb, Jamie Polley, Nikolai Yazgan, Melih Zöllner, J. Marius |
| contents | Realistic adversarial attacks on various camera-based perception tasks of autonomous vehicles have been successfully demonstrated so far. However, only a few works considered attacks on traffic light detectors. This work shows how CNNs for traffic light detection can be attacked with printed patches. We propose a threat model, where each instance of a traffic light is attacked with a patch placed under it, and describe a training strategy. We demonstrate successful adversarial patch attacks in universal settings. Our experiments show realistic targeted red-to-green label-flipping attacks and attacks on pictogram classification. Finally, we perform a real-world evaluation with printed patches and demonstrate attacks in the lab settings with a mobile traffic light for construction sites and in a test area with stationary traffic lights. Our code is available at https://github.com/KASTEL-MobilityLab/attacks-on-traffic-light-detection. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_04823 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Fool the Stoplight: Realistic Adversarial Patch Attacks on Traffic Light Detectors Pavlitska, Svetlana Robb, Jamie Polley, Nikolai Yazgan, Melih Zöllner, J. Marius Computer Vision and Pattern Recognition Machine Learning Realistic adversarial attacks on various camera-based perception tasks of autonomous vehicles have been successfully demonstrated so far. However, only a few works considered attacks on traffic light detectors. This work shows how CNNs for traffic light detection can be attacked with printed patches. We propose a threat model, where each instance of a traffic light is attacked with a patch placed under it, and describe a training strategy. We demonstrate successful adversarial patch attacks in universal settings. Our experiments show realistic targeted red-to-green label-flipping attacks and attacks on pictogram classification. Finally, we perform a real-world evaluation with printed patches and demonstrate attacks in the lab settings with a mobile traffic light for construction sites and in a test area with stationary traffic lights. Our code is available at https://github.com/KASTEL-MobilityLab/attacks-on-traffic-light-detection. |
| title | Fool the Stoplight: Realistic Adversarial Patch Attacks on Traffic Light Detectors |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2506.04823 |