Robust deepfake detection method based on siamese network

The proliferation of deepfake (DF) technology for generating manipulated facial expressions in synthetic images has raised concerns due to its potential negative impacts on individuals and society. In response to the need for robust detection, researchers have been developing methods to identify dee...

Full description

Saved in:
Bibliographic Details
Main Author: LIN Shanhe
Format: Article
Language:English
Published: POSTS&TELECOM PRESS Co., LTD 2024-04-01
Series:网络与信息安全学报
Subjects:
Online Access:http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2024016
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850091818583064576
author LIN Shanhe
author_facet LIN Shanhe
author_sort LIN Shanhe
collection DOAJ
description The proliferation of deepfake (DF) technology for generating manipulated facial expressions in synthetic images has raised concerns due to its potential negative impacts on individuals and society. In response to the need for robust detection, researchers have been developing methods to identify deepfakes. While current detection methods perform well on high-quality images, they often falter when confronted with low-quality or compressed images. This study focused on enhancing the robustness of deepfake detection methods to address these limitations. A novel approach leveraging a Siamese network was proposed, designed to learn common forgery features across both high-quality and low-quality images. This was achieved by trading off some of the high-quality image feature extraction capabilities to bolster the representational capacity for low-quality images. The proposed method demonstrated an average accuracy exceeding 90% across various datasets with different compression levels, surpassing several existing detection techniques. The simplicity, effectiveness, and adaptability of the proposed method to different backbone networks were further substantiated through ablation experiments.
format Article
id doaj-art-51fd9bbb73c5440bb1719bb32e3ed140
institution DOAJ
issn 2096-109X
language English
publishDate 2024-04-01
publisher POSTS&TELECOM PRESS Co., LTD
record_format Article
series 网络与信息安全学报
spelling doaj-art-51fd9bbb73c5440bb1719bb32e3ed1402025-08-20T02:42:17ZengPOSTS&TELECOM PRESS Co., LTD网络与信息安全学报2096-109X2024-04-011018218963897597Robust deepfake detection method based on siamese networkLIN ShanheThe proliferation of deepfake (DF) technology for generating manipulated facial expressions in synthetic images has raised concerns due to its potential negative impacts on individuals and society. In response to the need for robust detection, researchers have been developing methods to identify deepfakes. While current detection methods perform well on high-quality images, they often falter when confronted with low-quality or compressed images. This study focused on enhancing the robustness of deepfake detection methods to address these limitations. A novel approach leveraging a Siamese network was proposed, designed to learn common forgery features across both high-quality and low-quality images. This was achieved by trading off some of the high-quality image feature extraction capabilities to bolster the representational capacity for low-quality images. The proposed method demonstrated an average accuracy exceeding 90% across various datasets with different compression levels, surpassing several existing detection techniques. The simplicity, effectiveness, and adaptability of the proposed method to different backbone networks were further substantiated through ablation experiments.http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2024016forgery forensicsDeepfake detectionsiamese networkrobustness
spellingShingle LIN Shanhe
Robust deepfake detection method based on siamese network
网络与信息安全学报
forgery forensics
Deepfake detection
siamese network
robustness
title Robust deepfake detection method based on siamese network
title_full Robust deepfake detection method based on siamese network
title_fullStr Robust deepfake detection method based on siamese network
title_full_unstemmed Robust deepfake detection method based on siamese network
title_short Robust deepfake detection method based on siamese network
title_sort robust deepfake detection method based on siamese network
topic forgery forensics
Deepfake detection
siamese network
robustness
url http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2024016
work_keys_str_mv AT linshanhe robustdeepfakedetectionmethodbasedonsiamesenetwork