Article Detail

Ali Eren  Tuğrul*, Yakup Kutlu

Keywords

Deepfake Detection Transfer Learning Xception Generalization Out-of-Distribution DF40 EfficientNet ResNet

Doi : 10.71350/jere.2026178

Abstract

Deepfake technology, which enables the creation of highly realistic synthetic face images using deep learning, poses significant threats to information integrity and digital trust. This study presents a comparative evaluation of deep learning architectures for image-based deepfake detection, with a particular focus on generalization across different data distributions. Five models were implemented and evaluated: MesoNet, ResNet-50, EfficientNet-B4, Xception, and an extended Xception model retrained with an additional DF40-based external dataset. Experiments were conducted on the 140k Real and Fake Faces dataset, which contains 100,000 training, 20,000 validation, and 20,000 test images, and were extended with 25,696 external training images. All models were trained using transfer learning where applicable, CrossEntropyLoss with label smoothing, AdamW optimization, cosine annealing, mixed precision, early stopping, and checkpoint selection based on validation loss. The original Xception model achieved 99.73% accuracy on the in-distribution test set, but its external test accuracy dropped to 52.74%, with a fake recall of only 6.04%. In contrast, the Xception+DF40 model reached 92.15% accuracy, 0.9828 AUC, and 84.50% fake recall on the external test set while maintaining 99.10% accuracy on the original test set. These results show that data diversity and out-of-distribution evaluation are critical for robust deepfake detection.

References

1. Deng, J., Lin, C., Hu, P., Shen, C., Wang, Q., Li, Q., & Li, Q. (2024). Towards Benchmarking and Evaluating Deepfake Detection. IEEE Transactions on Dependable and Secure Computing. https://doi.org/10.1109/tdsc.2024.3369711
2. Dong, S., Wang, J., Ji, R., Liang, J., Fan, H., & Ge, Z. (2023). Implicit Identity Leakage: The Stumbling Block to Improving Deepfake Detection Generalization. 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). https://doi.org/10.1109/cvpr52729.2023.00389
3. Fang, S., Zhang, Z., & Song, B. (2025). Deepfake Detection Model Combining Texture Differences and Frequency Domain Information. ACM Transactions on Privacy and Security.
4. Gong, L. Y., & Li, X. J. (2024). A Contemporary Survey on Deepfake Detection: Datasets, Algorithms, and Challenges. Electronics. https://doi.org/10.3390/electronics13030585
5. Huang, P., Han, Y., Chu, E., Chen, J., & Hua, K. (2023). Multi-Task Self-Blended Images for Face Forgery Detection. ACM Multimedia Asia 2023. https://doi.org/10.1145/3595916.3626426
6. Kingra, S., Aggarwal, N., & Kaur, N. (2025). Assessing deepfake detection methods: a comparative evaluation on novel large-scale Asian deepfake dataset. International Journal of Data Science and Analytics.
7. Kumar, M., & Verma, B. (2026). Performance Evaluation of Face Forgery Detection Models: A Comparative Study. Lecture Notes in Networks and Systems.
8. Li, J., Xie, H., Yu, L., Gao, X., & Zhang, Y. (2021). Discriminative Feature Mining Based on Frequency Information and Metric Learning for Face Forgery Detection. IEEE Transactions on Knowledge and Data Engineering. https://doi.org/10.1109/tkde.2021.3117003
9. Luo, A., Cai, R., Kong, C., Ju, Y., Kang, X., Huang, J. (2024). Forgery-aware Adaptive Learning with Vision Transformer for Generalized Face Forgery Detection. IEEE Transactions on Circuits and Systems for Video Technology. https://doi.org/10.1109/tcsvt.2024.3522091
10. Ma, Z., Mei, X., & Shen, J. (2023). 3D Attention Network for Face Forgery Detection. 2023 4th Information Communication Technologies Conference (ICTC). https://doi.org/10.1109/ictc57116.2023.10154671
11. Peng, C., Chen, T., Liu, D., Guo, H., Wang, N., & Gao, X. (2025). Revisiting face forgery detection towards generalization. Neural Networks.
12. Ramachandran, S., Nadimpalli, A. V., & Rattani, A. (2021). An Experimental Evaluation on Deepfake Detection using Deep Face Recognition. 2021 International Carnahan Conference on Security Technology (ICCST). https://doi.org/10.1109/iccst49569.2021.9717407
13. Sun, R., Zhao, Z., Shen, L., Zeng, Z., Li, Y., Veeravalli, B. (2023). An Efficient Deep Video Model For Deepfake Detection. 2023 IEEE International Conference on Image Processing (ICIP). https://doi.org/10.1109/icip49359.2023.10222682
14. Tian, C., Luo, Z., Shi, G., & Li, S. (2023). Frequency-Aware Attentional Feature Fusion for Deepfake Detection. ICASSP 2023. https://doi.org/10.1109/icassp49357.2023.10094654
15. Zhuang, W., Chu, Q., Yuan, H., Miao, C., Liu, B., & Yu, N. (2022). Towards Intrinsic Common Discriminative Features Learning for Face Forgery Detection Using Adversarial Learning. 2022 IEEE International Conference on Multimedia and Expo (ICME). https://doi.org/10.1109/icme52920.2022.9859586