Citation

Chong, Jun Xiong and Goh, Michael Kah Ong and Tee, Connie (2022) Deepfakes Detection using Computer Vision and Deep Learning Approaches. Journal of System and Management Sciences, 12 (5). pp. 21-35. ISSN 1816-6075, 1818-0523

Text 11.pdf Restricted to Repository staff only Download (358kB)

Abstract

Earlier in 2018, deepfakes had grown in popularity as programmers used cutting-edge AI techniques to make software that could swap one person's face for another. The growth of deepfakes has not slowed down with each iteration of improvement and new approaches to swap faces. In 2019, Facebook, Tiktok, and Microsoft have started to block deepfakes videos and photos that might cause consumers to believe a subject act is from a real person. Humans' capacity to distinguish between face-swapped photos is no longer taken into account while trying to find a solution. In order to combat the false information that could harm some people, techniques to detect deepfakes are crucial. The goal of this research is to examine the most cutting-edge methods now available for identifying deepfake photos and to suggest a new or superior way utilizing computer vision and deep learning techniques. On the Face Forensic ++ DeepFake Dataset, the final models may achieve an Area Under the Curve (AUC) of 0.96661.

Item Type:	Article
Uncontrolled Keywords:	deepfakes, face augmentation, face detection, face manipulation, deep learning.
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	01 Dec 2022 03:50
Last Modified:	01 Dec 2022 03:50
URII:	http://shdl.mmu.edu.my/id/eprint/10878

Downloads

Edit (login required)