Comparative Evaluation of CNN and Transformer Architectures in Deepfake Detection Systems
International Journal of Advanced Scientific Research and Engineering Trends · 14.05.2026 · 10 (5) · 21-25
DOI: 10.65521/ijasret.v10i5.2838 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Advanced Scientific Research and Engineering Trends, online erschienen am 14.05.2026. Verfasst von Nilesh S. Kulkarni und 1 weiteren Autor:innen.
Abstract (Originalsprache)
The increase in generation of deepfake media which is also known as synthetic media has created ample challenges in verifying authenticity of digital content. In response, researchers have started working in this area for developing robust detection mechanisms. This paper offers a comprehensive comparative evaluation of Convolutional Neural Networks (CNNs) and Transformer-based architectures for deepfake detection. After taking insights from over 20 recent articles and papers, this analytical study examines detection accuracy, generalization ability, computational efficiency, and robustness of deepfake detection architectures. Results reveal that the CNNs effectively capture local features; Transformer models outperform them in modelling global dependencies, achieving superior detection accuracy across different datasets.
Zitieren
APA
Kulkarni, N. S., & Kharade, K. G. (2026). Comparative Evaluation of CNN and Transformer Architectures in Deepfake Detection Systems. International Journal of Advanced Scientific Research and Engineering Trends, 10(5), 21-25. https://doi.org/10.65521/ijasret.v10i5.2838
Vancouver
Kulkarni NS, Kharade KG. Comparative Evaluation of CNN and Transformer Architectures in Deepfake Detection Systems. International Journal of Advanced Scientific Research and Engineering Trends. 2026;10(5):21-25. doi:10.65521/ijasret.v10i5.2838
BibTeX
@article{Kulkarni2026,
author = {Kulkarni, Nilesh S. and Kharade, Kabir G.},
title = {Comparative Evaluation of CNN and Transformer Architectures in Deepfake Detection Systems},
journal = {International Journal of Advanced Scientific Research and Engineering Trends},
year = {2026},
volume = {10},
number = {5},
pages = {21--25},
doi = {10.65521/ijasret.v10i5.2838}
}
EUVIMED