Comparative Analysis of Deep-Fake Algorithms

Authors: Nikhil Sontakke, Sejal Utekar, Shivansh Rastogi, Shriraj Sonawane

Published: 2023-09-06 18:17:47+00:00

Comment: 7 pages, 4 figures, 2 tables, Published with International Journal of Computer Science Trends and Technology (IJCST)

Journal Ref: International Journal of Computer Science Trends and Technology (IJCST) V11(4): Page(109-115) Jul - Aug 2023. ISSN: 2347-8578

AI Summary

This paper provides a comprehensive review of the current state of deepfake creation and detection technologies. It examines various deep learning-based approaches used for both generating and identifying deepfakes, including their limitations and challenges. The review emphasizes the importance of continued research and development in this field to combat the negative impact of deepfakes on society and ensure the integrity of digital visual media.

Abstract

Due to the widespread use of smartphones with high-quality digital cameras and easy access to a wide range of software apps for recording, editing, and sharing videos and images, as well as the deep learning AI platforms, a new phenomenon of 'faking' videos has emerged. Deepfake algorithms can create fake images and videos that are virtually indistinguishable from authentic ones. Therefore, technologies that can detect and assess the integrity of digital visual media are crucial. Deepfakes, also known as deep learning-based fake videos, have become a major concern in recent years due to their ability to manipulate and alter images and videos in a way that is virtually indistinguishable from the original. These deepfake videos can be used for malicious purposes such as spreading misinformation, impersonating individuals, and creating fake news. Deepfake detection technologies use various approaches such as facial recognition, motion analysis, and audio-visual synchronization to identify and flag fake videos. However, the rapid advancement of deepfake technologies has made it increasingly difficult to detect these videos with high accuracy. In this paper, we aim to provide a comprehensive review of the current state of deepfake creation and detection technologies. We examine the various deep learning-based approaches used for creating deepfakes, as well as the techniques used for detecting them. Additionally, we analyze the limitations and challenges of current deepfake detection methods and discuss future research directions in this field. Overall, the paper highlights the importance of continued research and development in deepfake detection technologies in order to combat the negative impact of deepfakes on society and ensure the integrity of digital visual media.

Key findings

The paper's comparative analysis highlights various deep learning algorithms, including CNNs, RNNs, and LSTMs, as effective techniques for deepfake detection across different parameters. It concludes that while significant advancements have been made in identifying manipulated media, there is an ongoing need for continued research to counter evolving deepfake technologies. The review underscores the importance of developing more efficient models to enhance global security and reduce the spread of misinformation.

Approach

The paper conducts a comprehensive review of existing deepfake creation and detection methodologies. It describes deep learning-based generative models like GANs for creating deepfakes and analyzes various detection techniques, including CNNs, RNNs, and LSTMs, discussing their architectures and applications. Additionally, it surveys relevant datasets and compares the performance of different detection algorithms.

Datasets

Flickr-Faces-HQ (FFHQ), 100K-Faces, Diverse Fake Face Dataset (DFFD), CASIA-WebFace, VGGFace2, The Eye-Blinking Dataset, DeepfakeTIMIT

Model(s)

Generative Adversarial Networks (GANs), Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Long Short Term Memory (LSTM). Specific CNN variants mentioned include VGG19, VGG16, VGGFace, DenseNet169, DenseNet201, DenseNet121, and ResNet50.

Author countries

India

← Previous