EUVIMEDEUVIMED entdecken
Crossref

Deepfake Video Detection using Multi-Modality Features

J . Fahamitha, Sham Surya. R L, Siva Suriya R, Muniraja S, Thirumalai Srinivas B

International Journal of Advanced Research in Science Communication and Technology · 21.05.2026 · 97

DOI: 10.48175/ijarsct-35613 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Advanced Research in Science Communication and Technology, online erschienen am 21.05.2026. Verfasst von J . Fahamitha und 4 weiteren Autor:innen. Die Arbeit nennt 14 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The rapid advancement of Artificial Intelligence and Deep Learning technologies has revolutionized multimedia content creation, enabling the generation of highly realistic synthetic images, audio, and videos. Among these developments, deepfake videos have emerged as one of the most concerning threats to digital trust and cybersecurity. Deepfakes are artificially generated or manipulated videos in which a person’s face, voice, or actions are replaced with fabricated content using sophisticated neural networks such as Generative Adversarial Networks and autoencoders. Although these technologies have beneficial applications in entertainment and virtual media production, their misuse poses severe risks including misinformation campaigns, political manipulation, identity theft, financial fraud, and social instability. Traditional detection methods that rely solely on visual inspection or single-feature analysis are inadequate for identifying modern deepfakes because advanced synthesis techniques produce nearly perfect visual quality. To address this challenge, this project proposes a multi-modality deepfake detection framework that integrates visual features and Temporal-Oriented Information (TOI) spectrum analysis. Vision-based features capture spatial inconsistencies in facial textures and artifacts, while TOI spectrum features analyze temporal and frequency irregularities across frames. The fusion of these Experimental analysis demonstrates that the proposed approach significantly improves performance compared to single-modality systems. Therefore, the system provides an effective solution for ensuring digital media authenticity and combating deepfake threats

Zitieren

APA

Fahamitha, J., L, S. S. R., R, S. S., S, M., & B, T. S. (2026). Deepfake Video Detection using Multi-Modality Features. International Journal of Advanced Research in Science Communication and Technology, 97. https://doi.org/10.48175/ijarsct-35613

Vancouver

Fahamitha J, L SSR, R SS, S M, B TS. Deepfake Video Detection using Multi-Modality Features. International Journal of Advanced Research in Science Communication and Technology. 2026:97. doi:10.48175/ijarsct-35613

BibTeX
@article{Fahamitha2026,
  author = {Fahamitha, J . and L, Sham Surya. R and R, Siva Suriya and S, Muniraja and B, Thirumalai Srinivas},
  title = {Deepfake Video Detection using Multi-Modality Features},
  journal = {International Journal of Advanced Research in Science Communication and Technology},
  year = {2026},
  pages = {97},
  doi = {10.48175/ijarsct-35613}
}
Kostenlos starten →