NORMA eResearch @NCI Library

Comparative study of state of the art deepfake detection models

Kommalapati, Ajay Kumar (2022) Comparative study of state of the art deepfake detection models. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (653kB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (1MB) | Preview


Data analytics through object recognition and human-level control have all been effectively solved using deep learning. It's not all good news for privacy, democracy, and national security, though, because of developments in deep learning. Deepfake is a recent example of a deep learning powered application. Fake photos and movies created by deepfake algorithms might be difficult for people to tell apart from real ones. Digital visual media must consequently be able to automatically identify and analyse the integrity of its content. This paper studies the comparison between state of art with two online scanners such as WeVerify and Deepware. Using an EfficientNet backbone trained on ImageNet, these online scanners pretrain with the various datasets and employ an ensemble of five models. For the celeb-df dataset, DefakeHop achieved an AUC of 94.95 percent, while the online scanner Weverify and deepware achieved an accuracy rate of 97 and 75 respectively.

Item Type: Thesis (Masters)
Uncontrolled Keywords: AUC; Deepfake; EfficientNet; ImageNet; DefakeHop; Deepware
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
T Technology > TR Photography
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4150 Computer Network Resources > The Internet
T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Telecommunications > The Internet
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 21 Feb 2023 15:51
Last Modified: 02 Mar 2023 09:41

Actions (login required)

View Item View Item