A Comprehensive Survey of Attention-Based Deep Learning Techniques for Robust Deepfake Face Detection
Main Article Content
Abstract
The rapid advancement of deep learning and generative artificial intelligence has significantly increased the realism of artificially generated and manipulated facial images, creating substantial challenges for digital media authenticity and security. Fake-face detection has emerged as an important research area for identifying synthetic faces, face-swapping operations, and manipulated facial content. Traditional detection approaches often struggle to identify subtle manipulation artifacts, particularly under variations in image quality, compression, lighting conditions, and facial expressions. This paper presents a comprehensive survey of attention-based deep learning techniques for robust fake-face detection, focusing on recent advancements in convolutional neural networks (CNNs), vision transformers, and hybrid deep learning architectures. The survey examines the role of spatial attention, channel attention, self-attention, and multi-head attention mechanisms in enhancing discriminative feature extraction and identifying subtle inconsistencies in facial regions. Furthermore, it reviews attention-based feature fusion strategies, transfer learning approaches, and multi-scale feature extraction techniques used to improve detection performance. Various benchmark datasets, evaluation metrics, and experimental methodologies are discussed to provide a structured understanding of existing research. The paper also investigates key challenges, including cross-dataset generalization, adversarial manipulation, computational complexity, interpretability, and the detection of highly realistic AI-generated faces. Finally, potential research directions are identified, including complementary attention mechanisms, frequency-domain analysis, lightweight detection architectures, explainable artificial intelligence, and multimodal learning. This survey aims to provide researchers with a comprehensive understanding of existing attention-based detection methods and support the development of accurate, efficient, and generalizable fake-face detection frameworks.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
References
Y. Sheng, Z. Zou, Z. Yu, M. Pang, W. Ou, et al., “ID-insensitive deepfake detection model based on multi-attention mechanism,” Scientific Reports, vol. 15, Art. no. 11168, 2025, doi: 10.1038/s41598-025-96254-6.
S. Guo, Q. Li, M. Gao, X. Zhu, and I. Rida, “Generalizable deepfake detection via Spatial Kernel Selection and Halo Attention Network,” Image and Vision Computing, vol. 160, Art. no. 105582, 2025, doi: 10.1016/j.imavis.2025.105582.
A. P. Magesh, S. S. M. Ramakrishnan, R. A. Arun, N. Priyanka, and M. N. Kartheek, “Building an efficient Deep Fake detection system using the recognition capabilities of convolutional neural networks and transformers,” Discover Computing, vol. 28, Art. no. 99, 2025, doi: 10.1007/s10791-025-09586-2.
T. Zheng, “An attention-based framework for detecting face forgeries: Integrating Efficient-ViT and wavelet transform,” Mathematics, vol. 13, no. 16, Art. no. 2576, 2025, doi: 10.3390/math13162576.
J. E. Clapten and V. Balaji, “A gated temporal attention based intra prediction framework for robust deepfake video detection,” Scientific Reports, vol. 15, Art. no. 38540, 2025, doi: 10.1038/s41598-025-22333-3.
S. J. Shivaprakash, S. H., A. Chauhan, and A. Q. Md, “Attention-augmented hybrid framework with evolutionary optimization for robust deepfake detection,” Scientific Reports, vol. 16, Art. no. 24070, 2026, doi: 10.1038/s41598-026-51284-6.
M. Kumar, A. Kumar, V. Yadav, A. Shankar, and P. Chauhan, “A hybrid spatial–frequency attention-based algorithm using EfficientNet for robust and interpretable deepfake detection,” Scientific Reports, vol. 16, Art. no. 16612, 2026, doi: 10.1038/s41598-026-46086-9.
M. Javed, D. Han, F. H. Dahri, T. Kumar, J. A. Bhutto, et al., “Real time detection of deepfakes using the efficient Swin attention network with global and local facial features,” Discover Artificial Intelligence, vol. 6, Art. no. 508, 2026, doi: 10.1007/s44163-026-01188-1.
S. Gupta, O. Pandey, D. Varshney, M. Kumar, and G. C. Gandhi, “A hybrid attention-based spatio-temporal model for deepfake video detection,” Scientific Reports, 2026, doi: 10.1038/s41598-026-58264-w.
M. S. Reza, F. Elias, M. I. Mahmud, and N. Ahmed, “Attention and residual mechanism-based CNN architecture (ARC-Net) with enhanced fairness generalization for deepfake facial image detection,” PLOS ONE, vol. 21, no. 4, Art. no. e0340099, 2026, doi: 10.1371/journal.pone.0340099.
A. Bhattacharjee, K. Islam, K. Anan, A. Intesher, A. A. Fuad, U. Saha, and H. Imtiaz, “CAE-Net: Generalized deepfake image detection using convolution and attention mechanisms with spatial and frequency domain features,” Journal of Visual Communication and Image Representation, vol. 115, Art. no. 104679, 2026, doi: 10.1016/j.jvcir.2025.104679.
T. Rajesh and S. Maruthupermal, “Generalizable deepfake detection framework using hybrid convolution-based EfficientNetB7 with attention mechanism,” Expert Systems with Applications, vol. 297, Part B, Art. no. 129357, 2026, doi: 10.1016/j.eswa.2025.129357.
T. Hussain, B. T. Reddy, K. Phanindra, S. Terumalasetti, and G. A. Khan, “Decoding deception: State-of-the-art approaches to deep fake detection,” Frontiers in Big Data, vol. 8, Art. no. 1670833, 2026, doi: 10.3389/fdata.2025.1670833.
A. Raza, K. Munir, and M. Almutairi, “A novel deep learning approach for deepfake image detection,” Applied Sciences, vol. 12, no. 19, Art. no. 9820, 2022, doi: 10.3390/app12199820.
S. M. Zobaed, M. F. Rabby, M. I. Hossain, E. Hossain, S. Hasan, A. Karim, and K. M. Hasib, “DeepFakes: Detecting forged and synthetic media content using machine learning,” in Artificial Intelligence in Cyber Security: Impact and Implications. Cham, Switzerland: Springer, 2021, pp. 177–201.
V. Thambawita, J. L. Isaksen, S. A. Hicks, J. Ghouse, G. Ahlberg, A. Linneberg, N. Grarup, C. Ellervik, M. S. Olesen, T. Hansen, et al., “DeepFake electrocardiograms using generative adversarial networks are the beginning of the end for privacy issues in medicine,” Scientific Reports, vol. 11, Art. no. 21896, 2021, doi: 10.1038/s41598-021-01295-2.
M. F. B. Ahmed, M. S. U. Miah, A. Bhowmik, and J. B. Sulaiman, “Awareness to Deepfake: A resistance mechanism to Deepfake,” in Proc. 2021 International Congress of Advanced Technology and Engineering (ICOTEN), 2021, pp. 1–5, doi: 10.1109/ICOTEN52080.2021.9493549.
B. Dolhansky, R. Howes, B. Pflaum, N. Baram, and C. Canton Ferrer, “The Deepfake Detection Challenge (DFDC) Preview Dataset,” arXiv preprint arXiv:1910.08854, 2019.