AI Generated Image
AI-generated images are increasingly realistic, raising concerns about their misuse in misinformation and fraud. Current research focuses on developing robust detectors, often employing techniques like CLIP-based models, entropy-based methods, and Fourier analysis, to distinguish them from authentic photographs. These efforts are crucial for mitigating the societal impact of manipulated or fabricated visual content, and ongoing work emphasizes improving detector accuracy, generalization across different generative models, and understanding the underlying biases present in both generated images and detection models.
Papers
ASAP: Interpretable Analysis and Summarization of AI-generated Image Patterns at Scale
Jinbin Huang, Chen Chen, Aditi Mishra, Bum Chul Kwon, Zhicheng Liu, Chris Bryan
Which Model Generated This Image? A Model-Agnostic Approach for Origin Attribution
Fengyuan Liu, Haochen Luo, Yiming Li, Philip Torr, Jindong Gu