Deepfake Detection using AI – Manipulated Videos and Images
DOI:
https://doi.org/10.15662/IJARCST.2025.0805007Keywords:
Deepfake Detection, Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), CNN–LSTM Hybrid, Vision Transformers (ViTs), Generative Adversarial Networks (GANs), Grad-CAM, Preprocessing, Data Augmentation, Temporal Analysis, Lightweight ModelsAbstract
Artificial Intelligence (AI) has made it possible to develop deepfakes realistically manipulated videos and images, primarily created with Generative Adversarial Networks (GANs). Innovative though they are, their malicious use can lead to dangers in misinformation, political manipulation,fraud, and identity theft.This project introduces an AI-based deepfake detection system with Convolutional Neural Networks (CNNs),with XceptionNet being the baseline model, aided by preprocessing operations like face detection, frame extraction, and normalization. Trained on benchmark dataset including FaceForensics++ and DFDC, the system obtained 92.3% accuracy with robust precision, recall, and AUC-ROC values. Grad-CAM visualizations were incorporated for explainability, showing the regions manipulated. Even with issues including data imbalance, high computational intensity, and generalization concerns, the system was found to be scalable and effective. Improvements for the future include incorporating audio detection, real-time deployment, and transformer-based architectures towards higher resilience.
References
1. Dolhansky, B. et al. (2020). The Deepfake Detection Challenge (DFDC) Dataset. arXiv preprint arXiv:2006.07397.
2. Rossler, A. et al. (2019). FaceForensics++: Learning to Detect Manipulated Facial Images. arXiv preprint arXiv:1901.08971.
3. Chollet, F. (2017). Xception: Deep Learning with Depthwise Separable Convolutions. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
4. Goodfellow, I. et al. (2014). Generative Adversarial Networks (GANs). Advances in Neural Information Processing Systems (NeurIPS).
5. Verdoliva, L. (2020). Media forensics and deepfakes: an overview. IEEE Journal of Selected Topics in Signal Processing, 14(5), 910–932.
6. Perov, I. et al. (2020). DeepFaceLab: Integrated, flexible and extensible face-swapping framework. arXiv preprint arXiv:2005.05535.
7. Helmus, T. C. (2022). Artificial Intelligence, Deepfakes, and Disinformation. RAND Corporation.
8. Tolosana, R. et al. (2020). Deepfakes and Beyond: A Survey of Face Manipulation and Fake Detection. Information Fusion, 64, 131–148.
9. Zhang, C. et al. (2021). 3D Talking Face with Personalized Pose Dynamics. IEEE Transactions on Visualization and Computer Graphics, 29(2), 1438–1449.
10. Agarwal, S. et al. (2020). Detecting Deep-Fake Videos from Appearance and Behavior. IEEE WIFS.
