Fetching the paper…

Reading the bibliography…

2023

BadSAM: Exploring Security Vulnerabilities of SAM via Backdoor Attacks

Guan, Zihan, Hu, Mengxuan, Zhou, Zhongliang et al.

Original
PDF
Post on X

Understand

Recently, the Segment Anything Model (SAM) has gained significant attention as an image segmentation foundation model due to its strong performance on various downstream tasks.

However, it has been found that SAM does not always perform satisfactorily when faced with challenging downstream tasks.
This has led downstream users to demand a customized SAM model that can be adapted to these downstream tasks.
In this paper, we present BadSAM, the first backdoor attack on the image segmentation foundation model.

BadSAM: Exploring Security Vulnerabilities of SAM via Backdoor Attacks · Around

Built on

Anabranch network for camouflaged object segmentation
Trung-Nghia Le, Tam V Nguyen, Zhongliang Nie, Minh-Triet Tran, and Akihiro Sugimoto · 2019
Earlier work this paper cites.
Hidden backdoor attack against semantic segmentation models
Original
Yiming Li, Yanjie Li, Yalei Lv, Yong Jiang, and Shu-Tao Xia · 2021
Earlier work this paper cites.
Segment anything
Original
Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer Whitehead, Alexander C Berg, Wan-Yen Lo, et al · 2023
Earlier work this paper cites.

Similar

Text2seg: Remote sensing image semantic segmentation via text-guided visual foundation models
Original
Jielu Zhang, Zhongliang Zhou, Gengchen Mai, Lan Mu, Mengxuan Hu, and Sheng Li · 2023
Cited alongside, same era.
Segment anything model (sam) for digital pathology: Assess zero-shot segmentation on whole slide imaging
Original
Ruining Deng, Can Cui, Quan Liu, Tianyuan Yao, Lucas W Remedios, Shunxing Bao, Bennett A Landman, Lee E Wheless, Lori A Coburn, Keith T Wilson, et al · 2023
Cited alongside, same era.
Sam fails to segment anything?–sam-adapter: Adapting sam in underperformed scenes: Camouflage, shadow, and more
Original
Tianrun Chen, Lanyun Zhu, Chaotao Ding, Runlong Cao, Shangzhan Zhang, Yan Wang, Zejian Li, Lingyun Sun, Papa Mao, and Ying Zang · 2023
Cited alongside, same era.

Then

Influencer backdoor attack on semantic segmentation
Original
Haoheng Lan, Jindong Gu, Philip Torr, and Hengshuang Zhao · 2023
Closest in time.
Backdoor attacks to pre-trained unified foundation models
Original
Zenghui Yuan, Yixin Liu, Kai Zhang, Pan Zhou, and Lichao Sun · 2023
Closest in time.
Can sam segment anything? when sam meets camouflaged object detection
Original
Lv Tang, Haoke Xiao, and Bo Li · 2023
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…