Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2409.17874 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912047254470656 |
|---|---|
| author | Zhou, Ziqi Song, Yufei Li, Minghui Hu, Shengshan Wang, Xianlong Zhang, Leo Yu Yao, Dezhong Jin, Hai |
| author_facet | Zhou, Ziqi Song, Yufei Li, Minghui Hu, Shengshan Wang, Xianlong Zhang, Leo Yu Yao, Dezhong Jin, Hai |
| contents | Segment Anything Model (SAM) has recently gained much attention for its outstanding generalization to unseen data and tasks. Despite its promising prospect, the vulnerabilities of SAM, especially to universal adversarial perturbation (UAP) have not been thoroughly investigated yet. In this paper, we propose DarkSAM, the first prompt-free universal attack framework against SAM, including a semantic decoupling-based spatial attack and a texture distortion-based frequency attack. We first divide the output of SAM into foreground and background. Then, we design a shadow target strategy to obtain the semantic blueprint of the image as the attack target. DarkSAM is dedicated to fooling SAM by extracting and destroying crucial object features from images in both spatial and frequency domains. In the spatial domain, we disrupt the semantics of both the foreground and background in the image to confuse SAM. In the frequency domain, we further enhance the attack effectiveness by distorting the high-frequency components (i.e., texture information) of the image. Consequently, with a single UAP, DarkSAM renders SAM incapable of segmenting objects across diverse images with varying prompts. Experimental results on four datasets for SAM and its two variant models demonstrate the powerful attack capability and transferability of DarkSAM. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_17874 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | DarkSAM: Fooling Segment Anything Model to Segment Nothing Zhou, Ziqi Song, Yufei Li, Minghui Hu, Shengshan Wang, Xianlong Zhang, Leo Yu Yao, Dezhong Jin, Hai Artificial Intelligence Segment Anything Model (SAM) has recently gained much attention for its outstanding generalization to unseen data and tasks. Despite its promising prospect, the vulnerabilities of SAM, especially to universal adversarial perturbation (UAP) have not been thoroughly investigated yet. In this paper, we propose DarkSAM, the first prompt-free universal attack framework against SAM, including a semantic decoupling-based spatial attack and a texture distortion-based frequency attack. We first divide the output of SAM into foreground and background. Then, we design a shadow target strategy to obtain the semantic blueprint of the image as the attack target. DarkSAM is dedicated to fooling SAM by extracting and destroying crucial object features from images in both spatial and frequency domains. In the spatial domain, we disrupt the semantics of both the foreground and background in the image to confuse SAM. In the frequency domain, we further enhance the attack effectiveness by distorting the high-frequency components (i.e., texture information) of the image. Consequently, with a single UAP, DarkSAM renders SAM incapable of segmenting objects across diverse images with varying prompts. Experimental results on four datasets for SAM and its two variant models demonstrate the powerful attack capability and transferability of DarkSAM. |
| title | DarkSAM: Fooling Segment Anything Model to Segment Nothing |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2409.17874 |