The Creation and Detection of Deepfakes

By Yisroel Mirsky, Wenke Lee. ACM Computing Surveys, 2020

891 citations recorded by the source corpus.

Generative deep learning algorithms have progressed to a point where it is difficult to tell the difference between what is real and what is fake. In 2018, it was discovered how easy it is to use this technology for unethical and malicious applications, such as the spread of misinformation, impersonation of political leaders, and the defamation of innocent individuals. Since then, these “deepfakes” have advanced significantly. In this article, we explore the creation and detection of deepfakes and provide an in-depth view as to how these architectures work. The purpose of this survey is to provide the reader with a deeper understanding of (1) how deepfakes are created and detected, (2) the current trends and advancements in this domain, (3) the shortcomings of the current defense solutions, and (4) the areas that require further research and attention.

Subfields: Artificial Intelligence, Information Systems, Sociology and Political Science

View Paper: https://www.semanticscholar.org/paper/97ac037b7589d9d0f2cb5109aa2b095ad06067fd

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

In Sociology and Political Science

Citations in the corpus

Video Transformer for Deepfake Detection with Incremental Learning (2021, background): “Deep-fake can be grouped into four categories: face replacement, facial re-enactment, face editing, and complete face synthesis [37].”
Challenges in Deploying Machine Learning: A Survey of Case Studies (2020, background): “Closely related to this question is the growing concern of fake content being generated with ML, such as deepfake images and video, which can be easily used for the wrong purposes [95].”
DiffusionDB: A Large-scale Prompt Gallery Dataset for Text-to-Image Generative Models (2022, background): “focus on artifact detection and undirected detection (Mirsky and Lee, 2022).”
LION: Latent Point Diffusion Models for 3D Shape Generation (2022, background): “…negative use-cases for LION, it is important that practitioners apply an abundance of caution to mitigate impacts given generative modeling more generally can also be used for malicious purposes, discussed for instance in Vaccari and Chadwick [122], Nguyen et al. [123], Mirsky and Lee [124].”
Deepfake detection using deep learning methods: A systematic and comprehensive review (2023, methodology): “Mirsky and Lee (2021) specialized in human reenactment and replacement deepfakes.”
EditGAN: High-Precision Semantic Image Editing (2021, methodology): “As one potential way to tackle these challenges, methods for automatically validating real images and detecting manipulated or fake images are being developed by the research community [84, 85].”
Deepfake Video Detection Using Convolutional Vision Transformer (2021, methodology): “Some of the feature map representations are the Facial Action Coding System (FACS), image segmentation, facial landmarks, and facial boundaries [37].”
Generative artificial intelligence (2023, background): “…GAI models provides starting points for new applications and business models for misuse and criminals, ultimately being leveraged to spread misinformation and influence the media and politics, or to defraud individuals and businesses (Hart-mann et al., 2023; Kreps et al., 2022; Mirsky & Lee, 2022).”
Generative AI (2023, positive): “Notwithstanding, AI-generated disinformation has previously evolved as so-called deepfakes (Mirsky and Lee 2021), but recent advances in generative AI reduce the cost of creating such disinformation and allow for unprecedented personalization.”
Score-Based Generative Modeling with Critically-Damped Langevin Diffusion (2021, background): “Generative modeling of images has promising applications, for example for digital content creation and artistic expression (Bailey, 2020), but can also be used for nefarious purposes (Vaccari & Chadwick, 2020; Mirsky & Lee, 2021; Nguyen et al., 2021).”
Deep learning for deepfakes creation and detection: A survey (2019, background): “figures, virtually trying on clothes while shopping, and so on [19, 20].”
MACE: Mass Concept Erasure in Diffusion Models (2024, background): “These models can generate inappropriate content, such as copyrighted artworks [23, 24, 55, 61], explicit content [22, 59, 72], and deepfakes [38, 70].”