Deepfake Detection Using Convolutional Neural Network
International Journal of Innovative Science and Research Technology · 06.05.2026 · 3387
DOI: 10.38124/ijisrt/26apr2035 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Innovative Science and Research Technology, online erschienen am 06.05.2026. Verfasst von Anant N. Kaulage und 4 weiteren Autor:innen. Die Arbeit nennt 25 Literaturangaben.
Abstract (Originalsprache)
Deepfake technology, driven by modern AI models, has made video manipulation more convincing and harder to detect. This project aims to develop a deep learning based system that can automatically identify whether a video is real or artificially generated. The approach uses the EfficientNetB0 architecture as a feature extractor, trained on frames extracted from both authentic and manipulated video datasets. Frames were preprocessed, resized, and augmented to enhance the model’s learning ability to learn the given pattern. The training process involved two stages: initial training of top layers and subsequent fine-tuning of the entire network for higher accuracy. Further refinement was done using real video samples to enhance prediction stability. A threshold calibration method was applied to decide the real or fake nature of videos based on average prediction scores across frames. The final model efficiently distinguishes between real and fake content, demonstrating strong performance and potential use in video authenticity verification.
Zitieren
APA
Kaulage, A. N., Kolawale, P., Tuli, A., Liddad, I., & Jadhav, T. (2026). Deepfake Detection Using Convolutional Neural Network. International Journal of Innovative Science and Research Technology, 3387. https://doi.org/10.38124/ijisrt/26apr2035
Vancouver
Kaulage AN, Kolawale P, Tuli A, Liddad I, Jadhav T. Deepfake Detection Using Convolutional Neural Network. International Journal of Innovative Science and Research Technology. 2026:3387. doi:10.38124/ijisrt/26apr2035
BibTeX
@article{Kaulage2026,
author = {Kaulage, Anant N. and Kolawale, Pranjali and Tuli, Anika and Liddad, Isha and Jadhav, Tanvi},
title = {Deepfake Detection Using Convolutional Neural Network},
journal = {International Journal of Innovative Science and Research Technology},
year = {2026},
pages = {3387},
doi = {10.38124/ijisrt/26apr2035}
}
EUVIMED