CHRONICL OF DIPFEYK SCIENTIFIC RESEARCH: ANALYSIS
DOI:
https://doi.org/10.17721/2519-481X/2025/87-09Keywords:
deepfake, generative networks, multimodal detection, fake recognition, transformersAbstract
The article provides a systematic review of scientific research on the issue of deepfake detection for the period from 2015 to 2025. It demonstrates how the growth of generative capabilities (from early GANs to modern diffusion models) stimulated the development of detection tools. Key architectural solutions are analyzed: convolutional neural networks, transformers, multimodal models, and self-supervised approaches. Considerable attention is paid to open datasets that became the driving force of progress in this field: FaceForensics++, DFDC, Celeb-DF, ForgeryNet. A quantitative analysis of the dynamics of scientific publications based on Google Scholar data is carried out — in particular, a sharp increase in interest in the topic after 2018 is shown. An attempt is made to periodize technological development, presented in the form of a generalized trend table. Current challenges are also described: the emergence of realistic voice fakes, increasing complexity of generators, and attacks on detectors. The article includes a chart of the increase in the number of scientific works on the topic of deepfakes during 2015–2024, which demonstrates exponential growth. Special attention is paid to applied aspects: the use of fake media detection in cybersecurity, forensic examination, journalism, and personal data protection. The prospects for creating unified standards for digital content labeling, integrating recognition methods into real communication systems, and countering disinformation in social networks are considered. The importance of international coordination in solving the deepfake problem and the active involvement of the scientific community in the development of ethical principles for the use of AI is emphasized. All of this confirms the strategic relevance of the research topic not only today but also in the near future. The results can be used as a methodological basis for interdisciplinary initiatives.
References
1. Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A. and Bengio, Y., 2014. Generative adversarial nets. Advances in Neural Information Processing Systems, 27.
2. Li, Y., Chang, M. and Lyu, S., 2018. In ictu oculi: Exposing AI generated fake face videos by detecting eye blinking. arXiv preprint, arXiv:1806.02877. https://arxiv.org/abs/1806.02877
3. Korshunov, P. and Marcel, S., 2018. Deepfakes: a new threat to face recognition? Assessment and detection. arXiv preprint, arXiv:1812.08685. https://arxiv.org/abs/1812.08685
4. Afchar, D., Nozick, V., Yamagishi, J. and Echizen, I., 2018. MesoNet: A compact facial video forgery detection network. In: 2018 IEEE International Workshop on Information Forensics and Security (WIFS). IEEE.
5. Rössler, A., Cozzolino, D., Verdoliva, L., Riess, C., Thies, J. and Nießner, M., 2019. FaceForensics++: Learning to detect manipulated facial images. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). IEEE.
6. Dolhansky, B., Bitton, J., Pflaum, B., Lu, J., Howes, R., Wang, M. and Ferrer, C.C., 2019. The Deepfake Detection Challenge (DFDC) Preview Dataset. arXiv preprint, arXiv:1910.08854. https://arxiv.org/abs/1910.08854
7. Li, Y., Yang, X., Sun, P., Qi, H. and Lyu, S., 2020. Celeb-DF: A large-scale challenging dataset for deepfake forensics. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE.
8. He, Y., Zhang, H., Chang, W., Chen, X., Liu, Z., Sun, Q. and Lyu, S., 2021. ForgeryNet: A versatile benchmark for comprehensive forgery analysis. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE.
9. Zhang, D., Lin, F., Hua, Y., Wang, P., Zeng, D. and Ge, S., 2022. Deepfake Video Detection with Spatiotemporal Dropout Transformer. In: Proceedings of the 30th ACM International Conference on Multimedia.
10. Rahmouni, N., Nozick, V., Yamagishi, J. and Echizen, I., 2023. Compressed deepfake detection using a spatio-temporal approach. Procedia Computer Science, 219, pp. 436–444.
11. Nguyen, H.H., Yamagishi, J. and Echizen, I., 2023. Exploring self-supervised vision transformers for deepfake detection: a comparative analysis. arXiv preprint, arXiv:2305.00000.
12. Jain, A. and Bhushan, B., 2024. Ethics-aware artificial intelligence in the fight against deepfakes. AI and Ethics, 4(1), pp. 1–9.
