An Analysis of Recent Advances in Deepfake Image Detection in an Evolving Threat Landscape

arXiv:2404.16212v1 Announce Type: cross
Abstract: Deepfake or synthetic images produced using deep generative models pose serious risks to online platforms. This has triggered several research efforts to accurately detect deepfake images, achieving excellent performance on publicly available deepfake datasets. In this work, we study 8 state-of-the-art detectors and argue that they are far from being ready for deployment due to two recent developments. First, the emergence of lightweight methods to customize large generative models, can enable an attacker to create …

abstract advances analysis art arxiv cs.cr cs.cv cs.lg datasets deepfake deepfake images deep generative models detection generative generative models image image detection images landscape online platforms performance platforms research risks state study synthetic threat threat landscape type work