Real or Fake: Classification of AI Generated Synthetic Images

Dataset Source:

Our dataset was obtained from Kaggle via the link below:
https://www.kaggle.com/datasets/birdy654/cifake-real-and-ai-generated-synthetic-images

Models Used

We developed a VGG16, ResNet, and EfficientNet model for AI Image Classification. For the VGG16 model, in the training loop, the accuracy values are precentages.

Reference:

Krizhevsky, A., & Hinton, G. (2009). Learning multiple layers of features from tiny images
J. J. Bird and A. Lotfi, "CIFAKE: Image Classification and Explainable Identification of AI-Generated Synthetic Images," in IEEE Access, vol. 12, pp. 15642-15650, 2024, doi: 10.1109/ACCESS.2024.3356122. keywords: {Artificial intelligence;Visualization;Data models;Image recognition;Computational modeling;Synthetic data;Image classification;AI-generated images;generative AI;image classification;latent diffusion}