Reference : UNTAG: Learning Generic Features for Unsupervised Type-Agnostic Deepfake Detection
Scientific journals : Article
Engineering, computing & technology : Computer science
Computational Sciences
http://hdl.handle.net/10993/54532
UNTAG: Learning Generic Features for Unsupervised Type-Agnostic Deepfake Detection
English
Mejri, Nesryne mailto [University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > CVI2 >]
Ghorbel, Enjie mailto [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > CVI2 >]
Aouada, Djamila mailto [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > CVI2 >]
2023
IEEE International Conference on Acoustics, Speech and Signal Processing. Proceedings
IEEE. Institute of Electrical and Electronics Engineers
Yes
International
1520-6149
Greece
[en] Type-agnostic Deepfake Detection ; Unsupervised classification
[en] This paper introduces a novel framework for unsupervised type-agnostic deepfake detection called UNTAG. Existing methods are generally trained in a supervised manner at the classification level, focusing on detecting at most two types of forgeries; thus, limiting their generalization capability across different deepfake types. To handle that, we reformulate the deepfake detection problem as a one-class classification supported by a self-supervision mechanism. Our intuition is that by estimating the distribution of real data in a discriminative feature space, deepfakes can be detected as outliers regardless of their type. UNTAG involves two sequential steps. First, deep representations are learned based on a self-supervised pretext task focusing on manipulated regions. Second, a one-class classifier fitted on authentic image embeddings is used to detect deepfakes. The results reported on several datasets show the effectiveness of UNTAG and the relevance of the proposed new paradigm. The code is publicly available.
Interdisciplinary Centre for Security, Reliability and Trust (SnT) > CVI┬▓ - Computer Vision Imaging & Machine Intelligence
Fonds National de la Recherche - FnR
UNsupervised multi-type explainable deepFAKE detection (UNFAKE)
Researchers ; Professionals ; Students
http://hdl.handle.net/10993/54532
FnR ; FNR16763798 > Nesryne Mejri > UNFAKE > Unsupervised Multi-type Explainable Deepfake Detection > 01/09/2021 > 31/07/2025 > 2021

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Limited access
2023044110.pdfAuthor preprint1.97 MBRequest a copy

Bookmark and Share SFX Query

All documents in ORBilu are protected by a user license.