ANALYSIS OF DEEPFAKE VIDEO IDENTIFICATION METHODS
Authors/Creators
- 1. Department of Artificial Intelligence, Tashkent University of Information Technologies named after Muhammad al-Khwarizmi
- 2. Department of Computer engineering, University of Tashkent for Applied Sciences
Description
The rapid development of artificial intelligence and deep learning
technologies has significantly improved the quality of synthetic media, particularly
deepfake videos. While deepfakes have beneficial applications in entertainment,
education, and digital content creation, they also pose serious threats to privacy,
cybersecurity, journalism, and public trust. Consequently, the identification of
manipulated videos has become a critical research area. This paper analyzes the most
effective methods used for deepfake video detection, including visual artifact analysis,
temporal consistency analysis, biometric feature extraction, and deep learning-based
detection approaches. The study examines the strengths and limitations of current
techniques and highlights the challenges posed by increasingly sophisticated deepfake
generation models. The findings indicate that hybrid detection systems combining
multiple approaches achieve the highest detection accuracy and robustness
Files
79. Raxmanov M., Raxmanova O.- ANALYSIS OF DEEPFAKE VIDEO IDENTIFICATION METHODS.pdf
Additional details
Software
- Repository URL
- https://infocom.uz/magazine/26
- Development Status
- Active
References
- 1. Wang, Y., Zhao, H., & Li, X. (2024). Vision Transformer-Based Deepfake Video Detection with Spatial Attention Mechanisms. IEEE Transactions on Information Forensics and Security, 19, 2156–2169. 2. Li, J., Chen, S., Zhang, Y., & Xu, M. (2024). Spatiotemporal Deepfake Video Detection Using CNN-LSTM Hybrid Networks. Pattern Recognition, 149, 110245. 3. Zhang, Q., Wu, P., & Liu, H. (2024). Frequency-Aware Deepfake Detection via Spatial-Frequency Feature Fusion. Expert Systems with Applications, 248, 123654. 4. Huang, Z., Wang, R., & Sun, J. (2025). Detecting Diffusion-Based Deepfake Videos: Challenges and New Perspectives. Computer Vision and Image Understanding, 252, 104412. 5. Chen, X., Li, Y., & Zhou, F. (2025). Explainable Transformer Networks for Deepfake Video Identification. Knowledge-Based Systems, 312, 113214. 6. Singh, A., & Kumar, R. (2025). Self-Supervised Representation Learning for Generalized Deepfake Detection. Neural Networks, 186, 107154. 7. Liu, T., Yang, K., & Zhao, L. (2025). Multimodal Deepfake Detection through Audio-Visual Feature Fusion. IEEE Access, 13, 55231–55245. 8. Patel, D., Sharma, V., & Gupta, N. (2024). Robust Deepfake Detection under Video Compression Using Adaptive Preprocessing. Journal of Visual Communication and Image Representation, 99, 104067. 9. Garcia, M., Fernandez, J., & Ruiz, P. (2025). A Comprehensive Benchmark for Deepfake Video Detection across GAN and Diffusion Models. Information Fusion, 118, 102873.