Deepfake Forensics: Integration of Computer Vision and Unsupervised Learning Techniques for Video Authenticity Analysis
Deepfake Forensics: Integration of Computer Vision and Unsupervised Learning Techniques for Video Authenticity Analysis
DOI:
https://doi.org/10.51473/rcmos.v1i1.2023.1869Keywords:
Deepfake Forensics; Unsupervised Learning; Computer Vision; Autoencoders; Anomaly Detection; Cybersecurity; Zero-Generation.Abstract
The deepfake landscape demands a paradigm shift in detection, moving from Supervised Learning (SL) models that search for known artifacts to systems based on Unsupervised Learning (UL), capable of identifying anomalies and statistical deviations from authentic media. This paper provides a detailed analysis of integrating Advanced Computer Vision and UL Techniques to create robust Deepfake Forensics pipelines. The focus is on developing systems that do not depend on a predefined deepfake dataset, making them ideal for detecting zero-generationmanipulations or never-before-seen forgery techniques. We explore the use of Autoencoders and Generative Adversarial Networks (GANs) in their ability to model the distribution of normality (real videos) and the subsequent application of reconstruction metrics and latent mapping deviation to isolate the anomalous patterns that characterize the forgery. The practical application of these methodologies is vital for cybersecurity and content moderation on social media platforms, offering an authenticity verification mechanism that is resilient to the constant evolution of synthetic media technologies.
Downloads
References
AN, J.; CHO, S. Variational autoencoder based anomaly detection using reconstruction probability. Special Lecture on AI, 2015.
GOODFELLOW, I. et al. Generative adversarial networks. In: ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS. 2014.
LIDEN, L.; TULBURE, A.; VAN DER HALLEN, D. Deepfake detection using an anomaly detection approach. In: INTERNATIONAL CONFERENCE ON DIGITAL IMAGE PROCESSING. 2021.
SCHLEGEL, C. et al. f-AnoGAN: fast anomaly detection with GANs. In: MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION. 2019.
SUN, S. et al. Deepfake detection based on spatiotemporal features and unsupervised learning. In: IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING. 2020.
WANG, S. et al. Adversarial training for deepfake detection: a survey. arXiv, 2021.
XU, W. et al. Deep anomaly detection: a survey. arXiv, 2020.
YANG, X.; SUN, J.; ZHOU, S. Learning to detect digital forgeries with a weakly supervised approach. In: IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION. 2018.
ZHOU, P. et al. Two-stream neural networks for tampered face detection. In: IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS. 2017. DOI: https://doi.org/10.1109/CVPRW.2017.229
ROSSLER, A. et al. FaceForensics++: learning to detect manipulated facial images. In: INTERNATIONAL CONFERENCE ON COMPUTER VISION. 2019. DOI: https://doi.org/10.1109/ICCV.2019.00009
Downloads
Published
Issue
Section
Categories
License
Copyright (c) 2023 Matheus de Oliveira Pereira Paula (Autor)
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0). This means you are free to:
- Share — copy and redistribute the material in any medium or format.
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
The use of this material is conditioned upon proper attribution to the original author(s), providing a link to the license, and indicating if changes were made. The license does not require permission from the author or the publisher as long as these conditions are met.
The Creative Commons license logo is displayed permanently in the footer of the journal.
The copyright of the manuscript can be retained by the authors without restrictions and can be requested at any time, even after publication in the journal.
How to Cite
