Deepfake Detection: Comparison of Pretrained Xception and VGG16 Models

Nair, Karthika (2023) Deepfake Detection: Comparison of Pretrained Xception and VGG16 Models. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (2MB) | Preview

Abstract

The detrimental effects of deepfake on the society is the main topic of concern for this research. Deep learning is adopted to battle this issue as it is known for its exceptional proficiency in the effective detection of deepfakes by learning hierarchical features, intricate pattern and spatial relationships from the dataset. The emphasis is on employing CNNs especially Xception and VGG16 for addressing the threat of deepfake technology. The DFDC dataset is extracted from Kaggle aiding in utilising transfer learning to enhance the model efficiency. A comparison is carried out between employing transfer learning that works in collaboration with CNN as the baseline model and Xception and VGG16 are used for finetuning. The architectural design of this research includes the loading pretrained model, fine-tuning, and data augmentation, resulting in an effective system to detect deepfakes. The code execution is done by making use of python libraries such as Keras, Matplotlib, sklearn, Jupyter Notebook and Visual Studio code. The Xception model yields a remarkable accuracy of 84.6% whereas the VGG16 model gives an accuracy of 63%.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Subhnil, Shubham UNSPECIFIED
Uncontrolled Keywords:	Deepfake; Convolutional Neural Network (CNN); facial recognition; deep learning; transfer learning; data augmentation; Xception; VGG16
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence Q Science > QA Mathematics > Computer software > Computer Security T Technology > T Technology (General) > Information Technology > Computer software > Computer Security
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	18 May 2025 14:21
Last Modified:	18 May 2025 14:21
URI:	https://norma.ncirl.ie/id/eprint/7574

Actions (login required)

View Item