Repository logo
 
Publication

Comprehensive Evaluation of Deepfake Detection Models: Accuracy, Generalization, and Resilience to Adversarial Attacks

2025-01-25Text Open access
Simple item page
dc.contributor.authorAbbasi, Maryam
dc.contributor.authorANTUNES VAZ, PAULO JOAQUIM
dc.contributor.authorSilva, José
dc.contributor.authorMartins, Pedro
dc.date.accessioned2025-03-21T15:33:15Z
dc.date.available2025-03-21T15:33:15Z
dc.date.issued2025-01-25
dc.description.abstractThe rise of deepfakes—synthetic media generated using artificial intelli gence—threatens digital content authenticity, facilitating misinformation and manipu lation. However, deepfakes can also depict real or entirely fictitious individuals, leveraging state-of-the-art techniques such as generative adversarial networks (GANs) and emerging diffusion-based models. Existing detection methods face challenges with generalization across datasets and vulnerability to adversarial attacks. This study focuses on subsets of frames extracted from the DeepFake Detection Challenge (DFDC) and FaceForensics++ videos to evaluate three convolutional neural network architectures—XCeption, ResNet, and VGG16—for deepfake detection. Performance metrics include accuracy, precision, F1-score, AUC-ROC, and Matthews Correlation Coefficient (MCC), combined with an assessment of resilience to adversarial perturbations via the Fast Gradient Sign Method (FGSM). Among the tested models, XCeption achieves the highest accuracy (89.2% on DFDC), strong generalization, and real-time suitability, while VGG16 excels in precision and ResNet provides faster inference. However, all models exhibit reduced performance under adversarial conditions, underscoring the need for enhanced resilience. These find ings indicate that robust detection systems must consider advanced generative approaches, adversarial defenses, and cross-dataset adaptation to effectively counter evolving deep fake threatseng
dc.description.sponsorshipResearch Center in Digital Services (CISeD) and the Instituto Politécnico de Viseu for their support. FCT—Foundation for Science and Technology, I.P., through the institutional scientific employment program contract (CEECINST/00077/2021).
dc.identifier.citationAbbasi, M., Váz, P., Silva, J., & Martins, P. (2025). Comprehensive Evaluation of Deepfake Detection Models: Accuracy, Generalization, and Resilience to Adversarial Attacks. Applied Sciences, 15(3), 1225. https://doi.org/10.3390/app15031225
dc.identifier.doihttps:// doi.org/10.3390/app15031225
dc.identifier.eissn2076-3417
dc.identifier.urihttp://hdl.handle.net/10400.19/9293
dc.language.isoeng
dc.peerreviewedyes
dc.publisherMDPI
dc.relationCentre for Research in Digital Services
dc.relation.hasversionhttps://www.mdpi.com/2076-3417/15/3/1225
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.subject: deepfakes
dc.subjectdeep learning
dc.subjectXCeption
dc.subjectResNet
dc.subjectVGG
dc.subjectDFDC
dc.subjectFaceForensics++
dc.subjectadversarial robustness
dc.subjectdetection models
dc.titleComprehensive Evaluation of Deepfake Detection Models: Accuracy, Generalization, and Resilience to Adversarial Attackspor
dc.typetext
dspace.entity.typePublication
oaire.awardTitleCentre for Research in Digital Services
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F05583%2F2020/PT
oaire.citation.issue3
oaire.citation.startPage1225
oaire.citation.titleApplied Sciences
oaire.citation.volume15
oaire.fundingStream6817 - DCRRNI ID
oaire.versionhttp://purl.org/coar/version/c_970fb48d4fbd8a85
person.familyNameANTUNES VAZ
person.familyNameSilva
person.givenNamePAULO JOAQUIM
person.givenNameJosé
person.identifier.ciencia-id351C-9899-0EE7
person.identifier.ciencia-id4A14-D3E7-5B32
person.identifier.orcid0000-0002-1745-8937
person.identifier.orcid0000-0001-7285-8282
person.identifier.scopus-author-id55447844100
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.nameFundação para a Ciência e a Tecnologia
relation.isAuthorOfPublication702e79ee-5b0b-47ff-989d-12e6d8ea1e89
relation.isAuthorOfPublicatione9d8719e-af47-4008-b854-817801bb3964
relation.isAuthorOfPublication.latestForDiscovery702e79ee-5b0b-47ff-989d-12e6d8ea1e89
relation.isProjectOfPublicationa2335235-05b4-404c-a71e-cc37ae7fbf2c
relation.isProjectOfPublication.latestForDiscoverya2335235-05b4-404c-a71e-cc37ae7fbf2c

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Comprehensive Evaluation of Deepfake Detection Models_ Accuracy, Generalization, and Resilience to Adversarial Attacks.pdf
Size:
598.72 KB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.79 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

ESTGV - DEMGI - Artigo em revista científica, indexada ao WoS/Scopus
CISeD - Artigo em revista científica, indexada ao WoS/Scopus
ESTGV - DI - Artigo em revista científica, indexada ao WoS/Scopus