SegTrans: Transferable Adversarial Examples for Segmentation Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917001597812736 |
|---|---|
| author | Song, Yufei Zhou, Ziqi Lu, Qi Zhang, Hangtao Hu, Yifan Xue, Lulu Hu, Shengshan Li, Minghui Zhang, Leo Yu |
| author_facet | Song, Yufei Zhou, Ziqi Lu, Qi Zhang, Hangtao Hu, Yifan Xue, Lulu Hu, Shengshan Li, Minghui Zhang, Leo Yu |
| contents | Segmentation models exhibit significant vulnerability to adversarial examples in white-box settings, but existing adversarial attack methods often show poor transferability across different segmentation models. While some researchers have explored transfer-based adversarial attack (i.e., transfer attack) methods for segmentation models, the complex contextual dependencies within these models and the feature distribution gaps between surrogate and target models result in unsatisfactory transfer success rates. To address these issues, we propose SegTrans, a novel transfer attack framework that divides the input sample into multiple local regions and remaps their semantic information to generate diverse enhanced samples. These enhanced samples replace the original ones for perturbation optimization, thereby improving the transferability of adversarial examples across different segmentation models. Unlike existing methods, SegTrans only retains local semantic information from the original input, rather than using global semantic information to optimize perturbations. Extensive experiments on two benchmark datasets, PASCAL VOC and Cityscapes, four different segmentation models, and three backbone networks show that SegTrans significantly improves adversarial transfer success rates without introducing additional computational overhead. Compared to the current state-of-the-art methods, SegTrans achieves an average increase of 8.55% in transfer attack success rate and improves computational efficiency by more than 100%. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_08922 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SegTrans: Transferable Adversarial Examples for Segmentation Models Song, Yufei Zhou, Ziqi Lu, Qi Zhang, Hangtao Hu, Yifan Xue, Lulu Hu, Shengshan Li, Minghui Zhang, Leo Yu Computer Vision and Pattern Recognition Segmentation models exhibit significant vulnerability to adversarial examples in white-box settings, but existing adversarial attack methods often show poor transferability across different segmentation models. While some researchers have explored transfer-based adversarial attack (i.e., transfer attack) methods for segmentation models, the complex contextual dependencies within these models and the feature distribution gaps between surrogate and target models result in unsatisfactory transfer success rates. To address these issues, we propose SegTrans, a novel transfer attack framework that divides the input sample into multiple local regions and remaps their semantic information to generate diverse enhanced samples. These enhanced samples replace the original ones for perturbation optimization, thereby improving the transferability of adversarial examples across different segmentation models. Unlike existing methods, SegTrans only retains local semantic information from the original input, rather than using global semantic information to optimize perturbations. Extensive experiments on two benchmark datasets, PASCAL VOC and Cityscapes, four different segmentation models, and three backbone networks show that SegTrans significantly improves adversarial transfer success rates without introducing additional computational overhead. Compared to the current state-of-the-art methods, SegTrans achieves an average increase of 8.55% in transfer attack success rate and improves computational efficiency by more than 100%. |
| title | SegTrans: Transferable Adversarial Examples for Segmentation Models |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2510.08922 |