Proceedings of the International Conference on Intelligent Systems and Digital Transformation (ICISD 2025)

<Previous Article In Volume
Next Article In Volume>

Real-Time Deepfake Detection Using a Hybrid MobileNet-LSTM Model for Enhanced Media Integrity

Authors
S. Maheswari1, *, V. Dhilip Kumar1, D. Ajith Kumar1, R. Jahnavi1, C. Laharee1
1Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Department of Artificial Intelligence and Data science, Avadi, Chennai, Tamilnadu, India
*Corresponding author. Email: drmaheswaris@veltech.edu.in
Corresponding Author
S. Maheswari
Available Online 31 October 2025.
DOI
10.2991/978-94-6463-866-0_79How to use a DOI?
Keywords
Deepfake detection; MobileNet; LSTM; real-time analysis; Explainable AI; Grad-CAM; edge devices; scalable systems
Abstract

The proliferation of deepfake media stirred grave doubts over media authenticity threatening digital trust and social coherence. This paper proposes a new technique of deepfake detection from images and videos through the use of a hybrid of MobileNet spatial feature extraction and Long Short Term Memory (LSTM) networks for the analysis of temporal patterns. The proposed system ensures efficient real-time detection by achieving high classification accuracy while keeping computational efficiency. The model integrates Explainable AI (XAI) techniques such as Grad-CAM heatmaps to provide visual interpretation of the detected anomalies which can be measured using data from different the system is optimized for deployment on edge devices and offers scalability across platforms like social media news agencies and law enforcement of large scale applications. Future enhancements include extending system capabilities to multi-modal deepfake detection and continuous learning for evolving threats.

Copyright
© 2025 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

<Previous Article In Volume
Next Article In Volume>
Volume Title
Proceedings of the International Conference on Intelligent Systems and Digital Transformation (ICISD 2025)
Series
Atlantis Highlights in Intelligent Systems
Publication Date
31 October 2025
ISBN
978-94-6463-866-0
ISSN
2589-4919
DOI
10.2991/978-94-6463-866-0_79How to use a DOI?
Copyright
© 2025 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Cite this article

risenwbib
TY  - CONF
AU  - S. Maheswari
AU  - V. Dhilip Kumar
AU  - D. Ajith Kumar
AU  - R. Jahnavi
AU  - C. Laharee
PY  - 2025
DA  - 2025/10/31
TI  - Real-Time Deepfake Detection Using a Hybrid MobileNet-LSTM Model for Enhanced Media Integrity
BT  - Proceedings of the International Conference on Intelligent Systems and Digital Transformation (ICISD 2025)
PB  - Atlantis Press
SP  - 980
EP  - 991
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6463-866-0_79
DO  - 10.2991/978-94-6463-866-0_79
ID  - Maheswari2025
ER  -