Deepfake Video Traceability and Authentication via Source Attribution

In recent years, deepfake videos have emerged as a significant threat to societal and cybersecurity landscapes. Artificial intelligence (AI) techniques are used to create convincing deepfakes. The main counter method is deepfake detection. Currently, most of the mainstream detectors are based on dee...

Full description

Saved in:
Bibliographic Details
Main Authors: Canghai Shi, Minglei Qiao, Zhuang Li, Zahid Akhtar, Bin Wang, Meng Han, Tong Qiao
Format: Article
Language:English
Published: Wiley 2025-01-01
Series:IET Biometrics
Online Access:http://dx.doi.org/10.1049/bme2/5687970
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849717850516750336
author Canghai Shi
Minglei Qiao
Zhuang Li
Zahid Akhtar
Bin Wang
Meng Han
Tong Qiao
author_facet Canghai Shi
Minglei Qiao
Zhuang Li
Zahid Akhtar
Bin Wang
Meng Han
Tong Qiao
author_sort Canghai Shi
collection DOAJ
description In recent years, deepfake videos have emerged as a significant threat to societal and cybersecurity landscapes. Artificial intelligence (AI) techniques are used to create convincing deepfakes. The main counter method is deepfake detection. Currently, most of the mainstream detectors are based on deep neural networks. Such deep learning detection frameworks often face several problems that need to be addressed, for example, dependence on large-annotated datasets, lack of interpretability, and limited attention to source traceability. Towards overcoming these limitations, in this paper, we propose a novel training-free deepfake detection framework based on the interpretable inherent source attribution. The proposed framework not only distinguishes between real and fake videos but also traces their origins using camera fingerprints. Moreover, we have also constructed a new deepfake video dataset from 10 distinct camera devices. Experimental evaluations on multiple datasets show that the proposed method can attain high detection accuracies (ACCs) comparable to state-of-the-art (SOTA) deep learning techniques and also has superior traceability capabilities. This framework provides a robust and efficient solution for deepfake video authentication and source attribution, thus, making it highly adaptable to real-world scenarios.
format Article
id doaj-art-3e6432a5e2494e75b4b9317811d7698d
institution DOAJ
issn 2047-4946
language English
publishDate 2025-01-01
publisher Wiley
record_format Article
series IET Biometrics
spelling doaj-art-3e6432a5e2494e75b4b9317811d7698d2025-08-20T03:12:32ZengWileyIET Biometrics2047-49462025-01-01202510.1049/bme2/5687970Deepfake Video Traceability and Authentication via Source AttributionCanghai Shi0Minglei Qiao1Zhuang Li2Zahid Akhtar3Bin Wang4Meng Han5Tong Qiao6School of CyberspaceSchool of CyberspaceSchool of CyberspaceDepartment of Electrical and Computer EngineeringZhejiang Key Laboratory of Artificial Intelligence of Things (AIoT) Network and Data SecurityBinjiang Institute of Zhejiang UniversitySchool of CyberspaceIn recent years, deepfake videos have emerged as a significant threat to societal and cybersecurity landscapes. Artificial intelligence (AI) techniques are used to create convincing deepfakes. The main counter method is deepfake detection. Currently, most of the mainstream detectors are based on deep neural networks. Such deep learning detection frameworks often face several problems that need to be addressed, for example, dependence on large-annotated datasets, lack of interpretability, and limited attention to source traceability. Towards overcoming these limitations, in this paper, we propose a novel training-free deepfake detection framework based on the interpretable inherent source attribution. The proposed framework not only distinguishes between real and fake videos but also traces their origins using camera fingerprints. Moreover, we have also constructed a new deepfake video dataset from 10 distinct camera devices. Experimental evaluations on multiple datasets show that the proposed method can attain high detection accuracies (ACCs) comparable to state-of-the-art (SOTA) deep learning techniques and also has superior traceability capabilities. This framework provides a robust and efficient solution for deepfake video authentication and source attribution, thus, making it highly adaptable to real-world scenarios.http://dx.doi.org/10.1049/bme2/5687970
spellingShingle Canghai Shi
Minglei Qiao
Zhuang Li
Zahid Akhtar
Bin Wang
Meng Han
Tong Qiao
Deepfake Video Traceability and Authentication via Source Attribution
IET Biometrics
title Deepfake Video Traceability and Authentication via Source Attribution
title_full Deepfake Video Traceability and Authentication via Source Attribution
title_fullStr Deepfake Video Traceability and Authentication via Source Attribution
title_full_unstemmed Deepfake Video Traceability and Authentication via Source Attribution
title_short Deepfake Video Traceability and Authentication via Source Attribution
title_sort deepfake video traceability and authentication via source attribution
url http://dx.doi.org/10.1049/bme2/5687970
work_keys_str_mv AT canghaishi deepfakevideotraceabilityandauthenticationviasourceattribution
AT mingleiqiao deepfakevideotraceabilityandauthenticationviasourceattribution
AT zhuangli deepfakevideotraceabilityandauthenticationviasourceattribution
AT zahidakhtar deepfakevideotraceabilityandauthenticationviasourceattribution
AT binwang deepfakevideotraceabilityandauthenticationviasourceattribution
AT menghan deepfakevideotraceabilityandauthenticationviasourceattribution
AT tongqiao deepfakevideotraceabilityandauthenticationviasourceattribution