Copyright © 2025 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{180620,
        author = {Hamza Sayeed Quadri and Dr. P. Vishwapathi and Syed Nooruddin and Abdul Rahman},
        title = {Deepfake Detection Using CNN-LSTM Hybrid Model},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {1},
        pages = {1547-1551},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=180620},
        abstract = {The rapid growth of deep fake generative techniques is posing grooving challenges to media and information security. While there are methods to detect deep fakes today, they usually analyze spatial artifacts or issues in individual frames. This study proposes a temporal analysis framework using long short-term memory (LSTM) network to identify anomalies in facial movements. Our approach checks sequential patterns in facial features such as eye blinking and mouth movements. This proposed system was evaluated on deepfake data sets which was available to us through Kaggle, FaceForensics++ and CelebDF, and by employing preprocessing techniques like frame extraction and facial detection. If we compare this to traditional CNN-based methods, which solely rely on spatial features, LSTMs offer better detection accuracy by using temporal relationships. This project highlights the disadvantages of currently used detection system, such as generalization to overseen datasets and high computing complexities. The results achieved shows us a promising direction for real-time and efficient deepfake detection solutions.},
        keywords = {CNN-LSTM, Deepfake Detection, Temporal Analysis, Video Forensics},
        month = {June},
        }