UniShield: An Adaptive Multi-Agent Framework for Unified Forgery Image Detection and Localization
This work proposes UniShield, the novel multiagent-based unified system capable of detecting and localizing image forgeries across diverse domains, including image manipulation, document manipulation, DeepFake, and AI-generated images, which innovatively integrates a perception agent with a detection agent.
Qing Huang, Zhipei Xu, Xuan-Yu Zhang et al.
· 2026 IEEE/CVF Conference on... · 11 citations