ABSTRACT
Deepfake is an AI-based technology that makes videos that appear to be authentic but are actually fake, posing significant challenges to media credibility and public confidence. This study introduces a novel deepfake detection method that uses the Vision Transformer (ViT) model and is easily accessible via a web interface created using Streamlit. The technique is designed to differentiate between false and legitimate video footage with great efficiency, providing an effective tool for real-time content verification.
The inclusion of confidence scores increases the system’s interpretability, making it perfect for use in media authenticity and false news detection. The platform is simple to use, with plans to improve features such as the ability to automatically notify authorities, real-time video processing, the ability to work with multiple languages, and ways to share content via social media, all of which will contribute to its use in content protection.
Authors: Mathupriya S., Roopa D., Simon Jacob A., Santhosh G., Arvind M. | Department of Computer Science and Engineering Sri Sairam Institute of Technology Chennai, India.
Publisher: 2024 International Conference on Power, Energy, Control and Transmission Systems (ICPECTS) (IEEE Xplore)
Citation: M. S, R. D, S. J. A, S. G and A. M, “Deepfake Circumvention Using Machine Learning,” 2024 International Conference on Power, Energy, Control and Transmission Systems (ICPECTS), Chennai, India, 2024, pp. 1-6, doi: 10.1109/ICPECTS62210.2024.10780152.
References:
1.
Jeongho Kim, Taejune Kim, Jeonghyeon Kim and Simon S. Woo, “Evading Deepfake Detectors via High Quality Face Pre-Processing Methods ”, 26th International Conference on Pattern Recognition (ICPR) August 21-25, 2022, Montréal, Québec, Canada.
2.
Thokozile F. Mazibuko, Sudeep Tanwar, Pronaya Bhattacharya and Rajesh Gupta, “An Improved Dense CNN Architecture for Deepfake Image Detection ”, Received 14 January 2023, accepted 21 February 2023, date of publication 2 March 2023, date of current version 8 March 2023. Digital Object Identifier 10.1109/ACCESS.2023.3251417.
3.
Asad Malik, Minoru Kuribayashi, Sani M. Abdullahi and Ahmad Neyaz, “DeepFake Detection for Human Face Images and Videos: A Survey,” Received January 25, 2022, accepted February 9, 2022, date of publication February 11, 2022, date of current version February 22, 2022. Digital Object Identifier 10.1109/ACCESS.2022.3151186.
4.
MD Shohel Rana, Mohammad Nur Nobi, Beddhu Murali and Andrew H. Sung, “Deepfake Detection: A Systematic Literature Review,” Received January 25, 2022, accepted February 16, 2022, date of publication February 24, 2022, date of current version March 10, 2022.
5.
Amal Naitali, Mohammed Ridouani, Fatima Salahdine and Naima Kaabouch 3, “Deepfake Attacks: Generation, Detection, Datasets, Challenges, and Research Directions,” Computers 2023, 12, 216. https://doi.org/10.3390/computers12100216.
6.
Devanshu Shah, Devanshu Shah, Dhruvi Jodhawat, Jinay Parekh, Dr. Kriti Srivastava, “Xception Net & Vision Transformer: A comparative study for Deepfake Detection,” 2022 International Conference on Machine Learning, Computer Systems and Security (MLCSS)
7.
Hafsa Ilyas, Ali Javed, Khalid Mahmood Malik, and Aun Irtaza, “Deepfakes Examiner: An End-to-End Deep Learning Model for Deepfakes Videos Detection ”, 2022 16th International Conference on Open Source Systems and Technologies (ICOSST)
8.
Bismi Fathima Nasar, Sajini. T, and Elizabeth Rose Lalson, “A Survey on Deepfake Detection Techniques ”, International Journal of Computer Engineering in Research Trends, Survey Paper, Volume- 7, Issue- 8, August 2020 Refular Edition.
9.
Leandro A. Passosa, Danilo Jodasa, Kelton A. P. Costaa, Luis A. Souza Juniora, Douglas Rodriguesa, Javier Del Ser, David Camachod, Joao Paulo Papa, “A Review of Deep Learning-based Approaches for Deepfake Content Detection,” arXiv:2202.06095v2 [cs.CV] 10 Oct 2023.
10.
Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, Neil Houlsby, “An Image is Worth 16×16 Words: Transformation For Image Recognition At Scale,” Published as a conference paper at ICLR 2021.