Home/Journal of Computing and Communication/Article Record
Peer-Reviewed Article June 30, 2023

DeepFakeDG: A Deep Learning Approach for Deep Fake Detection and Generation

Are you an author of this work?Link to your sovereign AfriScholar profile
Published in: Journal of Computing and Communication
Publisher: Misr International University (MIU) • Egypt • 2636-3577
DOI: https://doi.org/10.21608/jocc.2023.307056

Abstract

The main idea of this project is to develop a web application that can help to detect whether the input data we provide of people, whether celebrities or people in general, are real or fake and generate deepfakes themselves. Recently, with the evolution of technology and advanced image editing tools, people can easily get manipulated, as deepfake algorithms can easily create fake videos and images that people can't distinguish from authentic ones, an emerging problem threatening the trustworthiness of online information. Deepfakes mainly affect public figures, celebrities, and politicians. Forged videos are videos that contain fake images over real ones. In this research, there are methods used with Machine and deep learning approaches that will be used with the dataset composed of deep fake videos and authentic ones to detect these manipulations and protect the government from criminals. There will be various techniques used to distinguish real from fake using face swapping, or is there something off regarding its behavior, or if a voice of a person is used with another person's voice, etc. The deep fake detector can be used in courts and police stations to reduce the likelihood of crimes and frauds that may happen and detect them. This project aims to make a website to detect whether videos are fake or not. More and above, the proposed model will also provide a deepfake generation efficiency.

Download Full Text PDFView at PublisherCrossref RecordCrossref XML

Verified Citation Velocity

Recorded scholarly citations in African and international peer-reviewed literature

14
Total Citations

Citations are continuously tracked and harvested across continental OJS endpoints and OpenAlex feeds.