A Dual-Branch Fusion Model for Deepfake Detection Using Video Frames and Microexpression Features

Deepfake detection has become a critical issue due to the rise of synthetic media and its potential for misuse. In this paper, we propose a novel approach to deepfake detection by combining video frame analysis with facial microexpression features. The dual-branch fusion model utilizes a 3D ResNet18...

Full description

Saved in:
Bibliographic Details
Main Authors: Georgios Petmezas, Vazgken Vanian, Manuel Pastor Rufete, Eleana E. I. Almaloglou, Dimitris Zarpalas
Format: Article
Language:English
Published: MDPI AG 2025-07-01
Series:Journal of Imaging
Subjects:
Online Access:https://www.mdpi.com/2313-433X/11/7/231
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850078018153742336
author Georgios Petmezas
Vazgken Vanian
Manuel Pastor Rufete
Eleana E. I. Almaloglou
Dimitris Zarpalas
author_facet Georgios Petmezas
Vazgken Vanian
Manuel Pastor Rufete
Eleana E. I. Almaloglou
Dimitris Zarpalas
author_sort Georgios Petmezas
collection DOAJ
description Deepfake detection has become a critical issue due to the rise of synthetic media and its potential for misuse. In this paper, we propose a novel approach to deepfake detection by combining video frame analysis with facial microexpression features. The dual-branch fusion model utilizes a 3D ResNet18 for spatiotemporal feature extraction and a transformer model to capture microexpression patterns, which are difficult to replicate in manipulated content. We evaluate the model on the widely used FaceForensics++ (FF++) dataset and demonstrate that our approach outperforms existing state-of-the-art methods, achieving 99.81% accuracy and a perfect ROC-AUC score of 100%. The proposed method highlights the importance of integrating diverse data sources for deepfake detection, addressing some of the current limitations of existing systems.
format Article
id doaj-art-28575c84ad014f1796a0a9ec3ad79cff
institution DOAJ
issn 2313-433X
language English
publishDate 2025-07-01
publisher MDPI AG
record_format Article
series Journal of Imaging
spelling doaj-art-28575c84ad014f1796a0a9ec3ad79cff2025-08-20T02:45:39ZengMDPI AGJournal of Imaging2313-433X2025-07-0111723110.3390/jimaging11070231A Dual-Branch Fusion Model for Deepfake Detection Using Video Frames and Microexpression FeaturesGeorgios Petmezas0Vazgken Vanian1Manuel Pastor Rufete2Eleana E. I. Almaloglou3Dimitris Zarpalas4Centre for Research and Technology Hellas, 57001 Thessaloniki, GreeceCentre for Research and Technology Hellas, 57001 Thessaloniki, GreeceHerta Security, 08037 Barcelona, SpainCentre for Research and Technology Hellas, 57001 Thessaloniki, GreeceCentre for Research and Technology Hellas, 57001 Thessaloniki, GreeceDeepfake detection has become a critical issue due to the rise of synthetic media and its potential for misuse. In this paper, we propose a novel approach to deepfake detection by combining video frame analysis with facial microexpression features. The dual-branch fusion model utilizes a 3D ResNet18 for spatiotemporal feature extraction and a transformer model to capture microexpression patterns, which are difficult to replicate in manipulated content. We evaluate the model on the widely used FaceForensics++ (FF++) dataset and demonstrate that our approach outperforms existing state-of-the-art methods, achieving 99.81% accuracy and a perfect ROC-AUC score of 100%. The proposed method highlights the importance of integrating diverse data sources for deepfake detection, addressing some of the current limitations of existing systems.https://www.mdpi.com/2313-433X/11/7/231deepfake detectionmicroexpressionsfusion model3D ResNettransformer
spellingShingle Georgios Petmezas
Vazgken Vanian
Manuel Pastor Rufete
Eleana E. I. Almaloglou
Dimitris Zarpalas
A Dual-Branch Fusion Model for Deepfake Detection Using Video Frames and Microexpression Features
Journal of Imaging
deepfake detection
microexpressions
fusion model
3D ResNet
transformer
title A Dual-Branch Fusion Model for Deepfake Detection Using Video Frames and Microexpression Features
title_full A Dual-Branch Fusion Model for Deepfake Detection Using Video Frames and Microexpression Features
title_fullStr A Dual-Branch Fusion Model for Deepfake Detection Using Video Frames and Microexpression Features
title_full_unstemmed A Dual-Branch Fusion Model for Deepfake Detection Using Video Frames and Microexpression Features
title_short A Dual-Branch Fusion Model for Deepfake Detection Using Video Frames and Microexpression Features
title_sort dual branch fusion model for deepfake detection using video frames and microexpression features
topic deepfake detection
microexpressions
fusion model
3D ResNet
transformer
url https://www.mdpi.com/2313-433X/11/7/231
work_keys_str_mv AT georgiospetmezas adualbranchfusionmodelfordeepfakedetectionusingvideoframesandmicroexpressionfeatures
AT vazgkenvanian adualbranchfusionmodelfordeepfakedetectionusingvideoframesandmicroexpressionfeatures
AT manuelpastorrufete adualbranchfusionmodelfordeepfakedetectionusingvideoframesandmicroexpressionfeatures
AT eleanaeialmaloglou adualbranchfusionmodelfordeepfakedetectionusingvideoframesandmicroexpressionfeatures
AT dimitriszarpalas adualbranchfusionmodelfordeepfakedetectionusingvideoframesandmicroexpressionfeatures
AT georgiospetmezas dualbranchfusionmodelfordeepfakedetectionusingvideoframesandmicroexpressionfeatures
AT vazgkenvanian dualbranchfusionmodelfordeepfakedetectionusingvideoframesandmicroexpressionfeatures
AT manuelpastorrufete dualbranchfusionmodelfordeepfakedetectionusingvideoframesandmicroexpressionfeatures
AT eleanaeialmaloglou dualbranchfusionmodelfordeepfakedetectionusingvideoframesandmicroexpressionfeatures
AT dimitriszarpalas dualbranchfusionmodelfordeepfakedetectionusingvideoframesandmicroexpressionfeatures