Transfer Learning of Real Image Features with Soft Contrastive Loss for Fake Image Detection

TL;DR

This paper introduces a transfer learning approach using soft contrastive loss to improve fake image detection by focusing on natural image traces, achieving high accuracy across diverse generative models and real-world scenarios.

Contribution

It proposes a novel self-supervised feature mapping and transfer learning method that enhances fake image detection by leveraging natural image traces and a soft contrastive loss.

Findings

Achieves 96.2% mAP on a diverse fake image dataset.

Surpasses baseline methods significantly in accuracy.

Demonstrates robustness across various transformations and real-world platforms.

Abstract

In the last few years, the artifact patterns in fake images synthesized by different generative models have been inconsistent, leading to the failure of previous research that relied on spotting subtle differences between real and fake. In our preliminary experiments, we find that the artifacts in fake images always change with the development of the generative model, while natural images exhibit stable statistical properties. In this paper, we employ natural traces shared only by real images as an additional target for a classifier. Specifically, we introduce a self-supervised feature mapping process for natural trace extraction and develop a transfer learning based on soft contrastive loss to bring them closer to real images and further away from fake ones. This motivates the detector to make decisions based on the proximity of images to the natural traces. To conduct a comprehensive experiment, we built a high-quality and diverse dataset that includes generative models comprising GANs and diffusion models, to evaluate the effectiveness in generalizing unknown forgery techniques and robustness in surviving different transformations. Experimental results show that our proposed method gives 96.2% mAP significantly outperforms the baselines. Extensive experiments conducted on popular commercial platforms reveal that our proposed method achieves an accuracy exceeding 78.4%, underscoring its practicality for real-world application deployment.