March 19, 2024, 4:49 a.m. | Yuxin Cao, Jinghao Li, Xi Xiao, Derui Wang, Minhui Xue, Hao Ge, Wei Liu, Guangwu Hu

cs.CV updates on arXiv.org arxiv.org

arXiv:2403.11656v1 Announce Type: new
Abstract: Previous work has shown that well-crafted adversarial perturbations can threaten the security of video recognition systems. Attackers can invade such models with a low query budget when the perturbations are semantic-invariant, such as StyleFool. Despite the query efficiency, the naturalness of the minutia areas still requires amelioration, since StyleFool leverages style transfer to all pixels in each frame. To close the gap, we propose LocalStyleFool, an improved black-box video adversarial attack that superimposes regional style-transfer-based …

abstract adversarial arxiv budget cs.cv efficiency low query recognition regional security segment segment anything segment anything model semantic style style transfer systems transfer type video work

Founding AI Engineer, Agents

@ Occam AI | New York

AI Engineer Intern, Agents

@ Occam AI | US

AI Research Scientist

@ Vara | Berlin, Germany and Remote

Data Architect

@ University of Texas at Austin | Austin, TX

Data ETL Engineer

@ University of Texas at Austin | Austin, TX

Business Intelligence Architect - Specialist

@ Eastman | Hyderabad, IN, 500 008