CoChat

Deepfake Detection: A Systematic Literature Review

Open access

The DOI matches its Crossref record; 150 of the 150 references it cites were found in a registry or an index.

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.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/access.2022.3154404

Cites

150 of 150 references verified0 citation errors · 0 not checkable

Statistics

Figures

FIGURE 1 Left: Google search engine finds web pages containing ''Deepfake'' keyword (web pages count vs. month). Right: Google search engine finds web pages holding Deepfake related videos (web pages count vs. month).
FIGURE 1 Left: Google search engine finds web pages containing ''Deepfake'' keyword (web pages count vs. month). Right: Google search engine finds web pages holding Deepfake related videos (web pages count vs. month).CC-BY · From the PDF
FIGURE 2 The process of the SLR.
FIGURE 2 The process of the SLR.CC-BY · From the PDF
TABLE 1. Define research questions for the SLR.
TABLE 1. Define research questions for the SLR.CC-BY · From the PDF
FIGURE 4 The information of the extracted data.
FIGURE 4 The information of the extracted data.CC-BY · From the PDF
FIGURE 5 Distribution of studies (half-yearly).
FIGURE 5 Distribution of studies (half-yearly).CC-BY · From the PDF
FIGURE 6 Steps of Deepfake detection.
FIGURE 6 Steps of Deepfake detection.CC-BY · From the PDF
FIGURE 7 List of datasets used in Deepfake related studies.
FIGURE 7 List of datasets used in Deepfake related studies.CC-BY · From the PDF
FIGURE 8 The list of Deepfake detection models.
FIGURE 8 The list of Deepfake detection models.CC-BY · From the PDF
FIGURE 9 The allocation of subcategories of detection models. ML: Machine Learning; DL: Deep Learning.
FIGURE 9 The allocation of subcategories of detection models. ML: Machine Learning; DL: Deep Learning.CC-BY · From the PDF
FIGURE 10 Taxonomy of Deepfake detection techniques. This taxonomy classifies the detection algorithms according to the media (image, video, or image and video), the features used (among the 12 features), the detection method (DL, ML, Blockchain, or statistical), and the clue for the detection (facial manipulation of digital media forensics, or other indications). The size of the connection line reflects the relative count of papers.
FIGURE 10 Taxonomy of Deepfake detection techniques. This taxonomy classifies the detection algorithms according to the media (image, video, or image and video), the features used (among the 12 features), the detection method (DL, ML, Blockchain, or statistical), and the clue for the detection (facial manipulation of digital media forensics, or other indications). The size of the connection line reflects the relative count of papers.CC-BY · From the PDF
FIGURE 11 The comparison of the results among deep learning and non-deep learning based models.
FIGURE 11 The comparison of the results among deep learning and non-deep learning based models.CC-BY · From the PDF
Image 12
CC-BY · From the PDF
Image 13
CC-BY · From the PDF
Image 14
CC-BY · From the PDF
Image 15
CC-BY · From the PDF

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

2 more claims about this paper

Create a free account to see every claim other papers make about this one.

See all claims — free account

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Research works printed but not linked: 13

  1. 10.1109/access.2020.3023037 · DOI · Paper page
  2. haliassos, 2020, Lips don’t lie: A generalisable and robust approach to face forgery detection, arXiv 2012 07657
  3. hui, 2021, How deep learning fakes videos (Deepfake) and how to detect it?
  4. oberoi, 2021, Exploring DeepFakes
  5. 2021, Contributing data to deepfake detection research
  6. 2021, Fakeapp
  7. 2021, FaceApp
  8. koopman, 2018, Detection of deepfake video manipulation, Proc 20th Irish Mach Vis image Process Conf (IMVIP), 1
  9. charitidis, 2020, Investigating the impact of pre-processing and prediction aggregation on the DeepFake detection task, arXiv 2006 07084
  10. matsakis, 2021, Artificial Intelligence is Now Fighting Fake Porn
  11. nhu, 2018, Forensics face detection from GANs using convolutional neural network, Proc Int Symp Inf Technol Converg (ISITC), 1
  12. tolosana, 2020, DeepFakes evolution: Analysis of facial regions and fake detection performance, arXiv 2004 07532
  13. 2020, IPFS powers the distributed web

Cited by

See all 811

Metadata

Authors
Rana, M. S., Nobi, M. N., Murali, B., & Sung, A. H.
Affiliations
  • Northern Kentucky University (US)
  • The University of Texas at San Antonio (US)
  • University of Southern Mississippi (US)
Venue
IEEE Access
Published
2022
Volume
10
Pages
25494-25513
ISSN
2169-3536 (ISSN-L)
OpenAlex
W4213449918
Cited by
573
References
135
Citations by year
2026: 141 · 2025: 207 · 2024: 155 · 2023: 61 · 2022: 9
FWCI
9.20
Type
journal article
Language
en
Open access
Locations
  • IEEE Access (publishedVersion, cc-by)
  • Aquila Digital Community (University of Southern Mississippi) (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • Northern Kentucky University
  • University of Southern Mississippi
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
  • Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
  • Physical Sciences › Mathematics › Statistics and Probability › Benford’s Law and Fraud Detection
Keywords
deepfake detection, deepfakes, video manipulation, misinformation, propaganda, classical machine learning, statistical techniques, detection performance evaluation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.