FakeFormer: Efficient Vulnerability-Driven Transformers for Generalisable Deepfake Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Nguyen, Dat, Astrid, Marcella, Ghorbel, Enjie, Aouada, Djamila
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909402693369856
author Nguyen, Dat
Astrid, Marcella
Ghorbel, Enjie
Aouada, Djamila
author_facet Nguyen, Dat
Astrid, Marcella
Ghorbel, Enjie
Aouada, Djamila
contents Recently, Vision Transformers (ViTs) have achieved unprecedented effectiveness in the general domain of image classification. Nonetheless, these models remain underexplored in the field of deepfake detection, given their lower performance as compared to Convolution Neural Networks (CNNs) in that specific context. In this paper, we start by investigating why plain ViT architectures exhibit a suboptimal performance when dealing with the detection of facial forgeries. Our analysis reveals that, as compared to CNNs, ViT struggles to model localized forgery artifacts that typically characterize deepfakes. Based on this observation, we propose a deepfake detection framework called FakeFormer, which extends ViTs to enforce the extraction of subtle inconsistency-prone information. For that purpose, an explicit attention learning guided by artifact-vulnerable patches and tailored to ViTs is introduced. Extensive experiments are conducted on diverse well-known datasets, including FF++, Celeb-DF, WildDeepfake, DFD, DFDCP, and DFDC. The results show that FakeFormer outperforms the state-of-the-art in terms of generalization and computational cost, without the need for large-scale training datasets. The code is available at \url{https://github.com/10Ring/FakeFormer}.
format Preprint
id arxiv_https___arxiv_org_abs_2410_21964
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle FakeFormer: Efficient Vulnerability-Driven Transformers for Generalisable Deepfake Detection
Nguyen, Dat
Astrid, Marcella
Ghorbel, Enjie
Aouada, Djamila
Computer Vision and Pattern Recognition
Recently, Vision Transformers (ViTs) have achieved unprecedented effectiveness in the general domain of image classification. Nonetheless, these models remain underexplored in the field of deepfake detection, given their lower performance as compared to Convolution Neural Networks (CNNs) in that specific context. In this paper, we start by investigating why plain ViT architectures exhibit a suboptimal performance when dealing with the detection of facial forgeries. Our analysis reveals that, as compared to CNNs, ViT struggles to model localized forgery artifacts that typically characterize deepfakes. Based on this observation, we propose a deepfake detection framework called FakeFormer, which extends ViTs to enforce the extraction of subtle inconsistency-prone information. For that purpose, an explicit attention learning guided by artifact-vulnerable patches and tailored to ViTs is introduced. Extensive experiments are conducted on diverse well-known datasets, including FF++, Celeb-DF, WildDeepfake, DFD, DFDCP, and DFDC. The results show that FakeFormer outperforms the state-of-the-art in terms of generalization and computational cost, without the need for large-scale training datasets. The code is available at \url{https://github.com/10Ring/FakeFormer}.
title FakeFormer: Efficient Vulnerability-Driven Transformers for Generalisable Deepfake Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.21964