Unmasking Deception: Deepfake detection using Shallow CNN

Singh, Ashwyn (2023) Unmasking Deception: Deepfake detection using Shallow CNN. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (2MB) | Preview
Preview	PDF (Configuration Manual) Download (994kB) | Preview

Abstract

Deepfakes are made using machine learning algorithms to create picture alteration and face swapping to present individuals speaking or doing things that never occurred. GANs have been used to generate fake pictures and videos because of its multipurpose approach. It utilizes ”landmark” points on the face to create such realistic fake videos which appear real to the naked eye until looked very carefully. Examples of such characteristics include the edges of an individuals eyes and lips, nostrils, and the contour of the jawline. These new advancements can negatively affect society, breaching privacy and security of individuals and organizations alike. The study utilizes shallow CNN along with post processing to identify deepfake videos efficiently. As deep learning is a field that is constantly developing, researching new techniques to detect these kind of videos is a must. According to the study done in this paper, the model performs well on an untrained dataset, with an accuracy of 86% percent and a logloss of 0.4084.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Horn, Christian UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software > Computer Security T Technology > T Technology (General) > Information Technology > Computer software > Computer Security Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	23 May 2025 10:09
Last Modified:	23 May 2025 10:09
URI:	https://norma.ncirl.ie/id/eprint/7615

Actions (login required)

View Item