Visual Deepfake Detection: Review of Techniques, Tools, Limitations, and Future Prospects
Open accessThe DOI matches its Crossref record; the references it cites are not checked yet.
Abstract
In recent years, rapid advancements in deepfakes (incorporating Artificial Intelligence (AI), machine, and deep learning) have updated tools and techniques for manipulating multimedia. Though technology has primarily been utilized for beneficial purposes, such as education and entertainment, it is also used for malicious or unethical tasks to spread disinformation or ruin someone’s dignity, even if it encompasses harassing and blackmailing victims. Deepfakes refer to high-quality and realistic multimedia-manipulated content that has been digitally modified or synthetically generated. We conducted a systematic literature review of deepfake detection to offer an updated overview of existing research work that initially describes the widely accessible deepfake generation tools, classifications, and detection process. We highlighted recent techniques in visual deepfake detection based on the feature representations, grouped into four domains: spatial, temporal, frequency, and spatio-temporal, including their key features and limitations by providing details of existing datasets, together with the potentials of deepfake and its future directions. This study tried to add an updated repository of technological change in deepfake, which could help researchers to develop robust deepfake models.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
In the citing paper
“high-fidelity deepfakes often minimize detectable arti -facts, and training deep models typically requires large, balanced datasets, which are not always available”
From Cross-dataset video deepfake detection using Transformer and CNN architectures (2026) · cites this paper as [6]
Citations
The citation lists are queued
They are fetched later and appear here then.
References 232 · Cited by 45
Research works printed but not linked: 66
- 2024, Artificial Intelligence is Now Fighting Fake Porn
- Lalla, 2024, Artificial Intelligence: Deepfakes in the Entertainment Industry
- Albahar, 2019, DeepFakes: Threats and countermeasures systematic review, J. Theor. Appl. Inf. Technol., 97, 3242
- 2024, Deepfake: Real Threat
- 2024, Defense Advanced Research Projects Agency
- Patil, 2023, Deepfake detection using biological features: A survey, arXiv:2301.05819 · arXiv
- Zhao, Dual-agent GANs for photorealistic and identity preserving profile face synthesis, 30, 1
- 2024, AI Tools
- Wang, 2024, Deep Face Swap With GAN
- 2024, Fewshot-Face-Translation-GAN
- 2024, DeepFaceLab 2.0
- 2024, Deepfaker App
- 2024, Deepfake TF
- Lattas, 2024, AvatarMe
- Ha, 2024, MarioNETte: Few-Shot Face Reenactment Preserving Identity of Unseen Targets
- 2024, Deepfakes Web
- 2024, DiscoFaceGAN
- 2024, Reface
- Tewari, 2024, StyleRig: Rigging StyleGAN for 3D Control Over Portrait Images
- 2024, Deepfake Studio
- Li, 2024, FaceShifter: Towards High Fidelity and Occlusion Aware Face Swapping
- 2024, Deepart
- Nirkin, 2024, FSGAN
- 2024, StyleGAN
- Thies, 2024, Face2Face: Real-Time Face Capture and Reenactment of RGB Videos
- 2024, Facemagic
- 2024, Faceapp
- Zhou, 2024, Neuraltexture
- Rampas, 2024, Instagram-Deepfake-Bot
- 2024, TB-Networks
- Chan, 2024, Everybody Dance Now
- Li, 2016, Deep identity-aware transfer of facial attributes, arXiv:1610.05586 · arXiv
- Lee, 2024, Deepfake Salvador Dalí Takes Selfies With Museum Visitors
- Dickson, 2019, Deepfake porn is still a threat, particularly for k-pop stars
- Huijstee, 2021, Tackling Deepfakes in European Policy
- 2024, Increasing Threats of Deepfake Identities
- Scott, 2024, Deepfakes a ’weapon Against Journalism,’ Analyst Says
- Castillo, 2024, Election Deepfakes Could Undermine Institutional Credibility, Moody’s Warns
- Zekić, Software quality assessment standards and metrics: A systematic literature review, 499
- Koopman, Detection of deepfake video manipulation, 133
- Prajapati, 2022, MRI-GAN: A generalized approach to detect DeepFakes using perceptual image assessment, arXiv:2203.00108 · arXiv
- Hong, Self-supervised deepfake detection by discovering artifact discrepancies, 1
- Cozzolino, 2018, ForensicTransfer: Weakly-supervised domain adaptation for forgery detection, arXiv:1812.02510 · arXiv
- Bounareli, 2022, Finding directions in GAN’s latent space for neural face reenactment, arXiv:2202.00046 · arXiv
- McCloskey, 2018, Detecting GAN-generated imagery using color cues, arXiv:1812.08247 · arXiv
- Alkishri, 2023, Deepfake image detection methods using discrete Fourier transform analysis and convolutional neural network, J. Jilin Univ. Eng. Technol. Ed., 42, 292
- Tan, 2024, Frequency-aware deepfake detection: Improving generalizability through frequency space learning, arXiv:2403.07240 · arXiv
- Khan, 2016, A survey on the state-of-the-art machine learning models in the context of NLP, Kuwait J. Sci., 43, 1
- Agarwal, Protecting world leaders against deep fakes, 38
- 10.1109/tpami.2020.3009287 · DOI · Paper page
- 10.1109/ICB45273.2019.8987375 · DOI · Paper page
- Kharel, 2023, DF-TransFusion: Multimodal deepfake detection via lip-audio cross-attention and facial self-attention, arXiv:2309.06511 · arXiv
- Guan, Delving into sequential patches for deepfake detection, 4517
- 10.7717/peerj-cs.1040 · DOI
- Guan, 2022, Detecting deepfake by creating spatio-temporal regularity disruption, arXiv:2207.10402 · arXiv
- Han, 2024, Towards more general video-based deepfake detection through facial feature guided adaptation for foundation model, arXiv:2404.05583 · arXiv
- 10.1109/tifs.2024.3409054 · DOI · Paper page
- Khalid, 2021, FakeAVCeleb: A novel audio-video multimodal deepfake dataset, arXiv:2108.05080 · arXiv
- Cai, 2023, AV-Deepfake1M: A large-scale LLM-driven audio-visual deepfake dataset, arXiv:2311.15308 · arXiv
- Bird, 2023, Real-time detection of AI-generated speech for DeepFake voice conversion, arXiv:2308.12734 · arXiv
- Yurk, 2024, The Next ’Deep Fake’ Images: Public Lands, Natural Disasters
- Al-Mulla, 2022, Deepfakes: Criminalization and legalization analytical descriptive study, Webology, 19
- 2024, Imaging the Internet| A History and Forecast
- Cyber Risks Associated With Generative Artificial Intelligence
- 2024, Deepfakes for Good? How Synthetic Media is Transforming Business
- 10.1038/s41551-021-00751-8 · DOI · Paper page
Metadata
- Authors
- Ur Rehman Ahmed, N., Badshah, A., Adeel, H., Tajammul, A., Daud, A., & Alsahfi, T.
- Affiliations
- Department of Computing, Hamdard University, Islamabad Campus, Islamabad, Pakistan
- University of Sargodha (PK)
- Mehran University of Engineering and Technology (PK)
- Rabdan Academy (AE)
- University of Jeddah (SA)
- Venue
- IEEE Access
- Published
- 2025
- Volume
- 13
- Pages
- 1923-1961
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W4405838418
- Cited by
- 45
- References
- 232
- Citations by year
- 2026: 26 · 2025: 19
- FWCI
- 6.25
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- IEEE Access (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Rabdan Academy Digital Repository (Rabdan Academy) (submittedVersion, cc-by)
- 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 › Computer Science › Computer Vision and Pattern Recognition › Generative Adversarial Networks and Image Synthesis
- Keywords
- deepfake detection, disinformation, deepfake datasets, feature representation, spatio-temporal features, temporal features, frequency characteristics
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Visual Deepfake Detection: Review of Techniques, Tools, Limitations, and Future Prospects. https://app.cochat.ai/papers/10.1109/access.2024.3523288
- The paper
- Ur Rehman Ahmed, N., Badshah, A., Adeel, H., Tajammul, A., Daud, A., & Alsahfi, T. (2025). Visual Deepfake Detection: Review of Techniques, Tools, Limitations, and Future Prospects. IEEE Access. https://doi.org/10.1109/access.2024.3523288
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.