Deepfake Detection: A Systematic Literature Review

Abstract

Over the last few decades, rapid progress in AI, machine learning, and deep learning has resulted in new techniques and various tools for manipulating multimedia. Though the technology has been mostly used in legitimate applications such as for entertainment and education, etc., malicious users have also exploited them for unlawful or nefarious purposes. For example, high-quality and realistic fake videos, images, or audios have been created to spread misinformation and propaganda, foment political discord and hate, or even harass and blackmail people. The manipulated, high-quality and realistic videos have become known recently as Deepfake. Various approaches have since been described in the literature to deal with the problems raised by Deepfake. To provide an updated overview of the research works in Deepfake detection, we conduct a systematic literature review (SLR) in this paper, summarizing 112 relevant articles from 2018 to 2020 that presented a variety of methodologies. We analyze them by grouping them into four different categories: deep learning-based techniques, classical machine learning-based methods, statistical techniques, and blockchain-based techniques. We also evaluate the performance of the detection capability of the various methods with respect to different datasets and conclude that the deep learning-based methods outperform other methods in Deepfake detection.

DFFMD: A Deepfake Face Mask Dataset for Infectious Disease Era With Deepfake Detection Algorithms

“…tive comparison to analyze the State-of-the-Art (SOTA) video Deepfake detection methods with our results based on the used measures and datasets in Table4. As confirmed by this study[49], the most used metrics in Deepfake are accuracy, an area under the ROC curve (AUC) (see Fig.15), and F1-score (see VOLUME ..., 2020…”

Towards a Positioning Model for Evaluating the Use and Design of Anti-Disinformation Tools

“…Lately, we have also witnessed the appearance of new forms of disinformation, which do not employ only written articles. One example of this typology of content is deepfakes, which are highly realistic videos that can manipulate the movements and voices of actors present in them to make them resemble realistic videos [40]. Examples of deepfakes have involved highly influential figures such as Barack Obama, Donald Trump, as well as journalists delivering news or tech moguls announcing technological innovations.…”

“…For these reasons and for the threat that deepfakes pose, research on the topic has been trying to understand the phenomenon and how to counter it. For example, Rana and colleagues [40] conducted a review of the current deepfake research, and Deshmukh and Wankhade [11] conducted a review of current deepfake detection technologies.…”

Missing the Bigger Picture: The Need for More Research on Visual Health Misinformation

“…Tools leveraging computer vision to automatically detect and classify visual content within these larger information ecosystems are less frequently deployed. Considerable research within computer science has explored how to detect manipulated or fabricated visual content (Rana et al, 2022). Combining the ability to detect visual misinformation with existing computational surveillance approaches would support the early identification of visual health misinformation and make automated identification of this content more effective (Singh et al, 2021).…”