DeepExplain: Enhancing DeepFake Detection Through Transparent and Explainable AI model
DOI:
https://doi.org/10.31449/inf.v48i8.5792Abstract
The rapid advancement of digital media technologies has given rise to DeepFake videos, synthetic content generated using deep learning algorithms that can convincingly mimic real individuals' appearances and actions. This presents a significant challenge in maintaining digital content integrity, as DeepFakes pose threats to information veracity, security, and public trust. Addressing this challenge necessitates robust detection methods that not only accurately identify DeepFakes but also ensure transparency and understandability in their operations. This study introduces DeepExplain, a new approach that combine convolutional neural networks (CNNs) and long shortterm memory (LSTM) networks, augmented with explainability features to enhance the detection of DeepFakes. Utilizing the comprehensive DeepFake Detection Challenge (DFDC) dataset, DeepExplain demonstrates superior accuracy and balanced performance across essential metrics such as recall, precision, and area under the receiveroperating characteristic (AUROC) curve. Crucially, by integrating explainability mechanisms like Gradientweighted Class Activation Mapping (Grad-CAM) and SHapley Additive exPlanations (SHAP) values, DeepExplain not only identifies DeepFakes but also provides insights into the decision-making process, fostering trust and facilitating broader understanding. The findings underscore the potential of explainable and transparent AI solutions in combating the evolving threat of DeepFakes, highlighting important directions for future research andpractical applications in digital media verification and securityReferences
W. H. Abir et al., "Detecting Deepfake Images Using Deep Learning Techniques and Explainable AI Methods".
Y. Mirsky and W. Lee, "The Creation and Detection of Deepfakes: A Survey".
J. Pu et al.. "Deepfake Videos in the Wild: Analysis and Detection". Apr. 2021. 10.1145/3442381.3449978.
S. Lyu. "DeepFake Detection: Current Challenges and Next Steps". arXiv (Cornell University). Mar. 2020. 10.48550/arxiv.2003.09234.
S. Lyu, "Deepfake Detection: Current Challenges and Next Steps".
P. J. Phillips, C. A. Hahn, P. Fontana, D. A. Broniatowski and M. A. Przybocki. "Four Principles of Explainable Artificial Intelligence". Aug. 2020. 10.6028/nist.ir.8312-draft.
S. Mohseni, N. Zarei and E. D. Ragan. "A Multidisciplinary Survey and Framework for Design and Evaluation of Explainable AI Systems". arXiv (Cornell University). Nov. 2018. 10.48550/arxiv.1811.11839.
A. Kiseleva, D. Kotzinos and P. D. Hert. "Transparency of AI in Healthcare as a Multilayered System of Accountabilities: Between Legal Requirements and Technical Limitations". Frontiers in artificial intelligence. vol. 5. May. 2022. 10.3389/frai.2022.879603.
Y. Mirsky and W. Lee. "The Creation and Detection of Deepfakes". ACM Computing Surveys. vol. 54. no. 1. pp. 1-41. Jan. 2021. 10.1145/3425780.
D. M. Montserrat et al.. "Deepfakes Detection with Automatic Face Weighting". Jun. 2020. 10.1109/cvprw50498.2020.00342.
C. Leibowicz, S. McGregor and A. Ovadya. "The Deepfake Detection Dilemma". Jul. 2021. 10.1145/3461702.3462584.
A. Singh, A. S. Saimbhi, N. Singh and M. Mittal. "DeepFake Video Detection: A Time-Distributed Approach". SN Computer Science. vol. 1. no. 4. Jun. 2020. 10.1007/s42979-020-00225-9.
C. Leibowicz, S. McGregor and A. Ovadya. "The Deepfake Detection Dilemma: A Multistakeholder Exploration of Adversarial Dynamics in Synthetic Media". Feb. 2021. 10.48550/arXiv.2102.06109v1.
"Deepfake detection by human crowds, machines, and machine-informed crowds | Proceedings of the National Academy of Sciences".
S. Wang, O. Wang, R. Zhang, A. Owens and A. A. Efros. "CNN-generated images are surprisingly easy to spot... for now". arXiv (Cornell University). Dec. 2019. 10.48550/arxiv.1912.11035.
T. Hwang, "Deepfakes: A Grounded Threat Assessment".
N. N. Thaw, T. July, A. N. Wai, D. H. Goh and A. Y. K. Chua. "Is it real? A study on detecting deepfake videos". Proceedings of the Association for Information Science and Technology. vol. 57. no. 1. Oct. 2020. 10.1002/pra2.366.
Y. Li, P. Sun, H. Qi and S. Lyu, "Toward the Creation and Obstruction of DeepFakes".
J. He, S. L. Baxter, J. Xu, J. Xu, X. Zhou and K. Zhang. "The practical implementation of artificial intelligence technologies in medicine". Nature Medicine. vol. 25. no. 1. pp. 30-36. Jan. 2019. 10.1038/s41591-018-0307-0.
S. Liu et al.. "Explainable Deep Learning for Uncovering Actionable Scientific Insights for Materials Discovery and Design". arXiv (Cornell University). Jul. 2020. 10.48550/arxiv.2007.08631.
W. Samek, T. Wiegand and K. Müller. "Explainable Artificial Intelligence: Understanding, Visualizing and Interpreting Deep Learning Models". arXiv (Cornell University). Aug. 2017.
A. Singh, S. Sengupta and V. Lakshminarayanan. "Explainable Deep Learning Models in Medical Image Analysis". Journal of Imaging. vol. 6. no. 6. pp. 52-52. Jun. 2020. 10.3390/jimaging6060052.
Downloads
Published
How to Cite
Issue
Section
License
I assign to Informatica, An International Journal of Computing and Informatics ("Journal") the copyright in the manuscript identified above and any additional material (figures, tables, illustrations, software or other information intended for publication) submitted as part of or as a supplement to the manuscript ("Paper") in all forms and media throughout the world, in all languages, for the full term of copyright, effective when and if the article is accepted for publication. This transfer includes the right to reproduce and/or to distribute the Paper to other journals or digital libraries in electronic and online forms and systems.
I understand that I retain the rights to use the pre-prints, off-prints, accepted manuscript and published journal Paper for personal use, scholarly purposes and internal institutional use.
In certain cases, I can ask for retaining the publishing rights of the Paper. The Journal can permit or deny the request for publishing rights, to which I fully agree.
I declare that the submitted Paper is original, has been written by the stated authors and has not been published elsewhere nor is currently being considered for publication by any other journal and will not be submitted for such review while under review by this Journal. The Paper contains no material that violates proprietary rights of any other person or entity. I have obtained written permission from copyright owners for any excerpts from copyrighted works that are included and have credited the sources in my article. I have informed the co-author(s) of the terms of this publishing agreement.
Copyright © Slovenian Society Informatika
