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

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liang, Ziyou, Liu, Weifeng, Wang, Run, Wu, Mengjie, Li, Boheng, Zhang, Yuyang, Wang, Lina, Yang, Xinyi
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909521756028928
author Liang, Ziyou
Liu, Weifeng
Wang, Run
Wu, Mengjie
Li, Boheng
Zhang, Yuyang
Wang, Lina
Yang, Xinyi
author_facet Liang, Ziyou
Liu, Weifeng
Wang, Run
Wu, Mengjie
Li, Boheng
Zhang, Yuyang
Wang, Lina
Yang, Xinyi
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2403_16513
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Transfer Learning of Real Image Features with Soft Contrastive Loss for Fake Image Detection
Liang, Ziyou
Liu, Weifeng
Wang, Run
Wu, Mengjie
Li, Boheng
Zhang, Yuyang
Wang, Lina
Yang, Xinyi
Computer Vision and Pattern Recognition
Cryptography and Security
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.
title Transfer Learning of Real Image Features with Soft Contrastive Loss for Fake Image Detection
topic Computer Vision and Pattern Recognition
Cryptography and Security
url https://arxiv.org/abs/2403.16513