Channel-Aware Local–Global Representation Learning for Generalizable Deepfake Detection
Mathematics · 01.06.2026 · 14 (11) · 1913
DOI: 10.3390/math14111913 ↗Kurz gefasst
Fachartikel in der Zeitschrift Mathematics, online erschienen am 01.06.2026. Verfasst von Liang Kuang und 2 weiteren Autor:innen. Die Arbeit nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Wuxi Science and Technology Innovation Fund “Taihu Light” Science and Technology Tackling Program, the Jiangsu Provincial Higher Vocational Colleges Engineering Technology Research and Development Center, the Jiangsu Provincial University Excellent Scientific and Technological Innovation Team.
Abstract (Originalsprache)
Deepfake detection has emerged as a critical research area in image content security. To address the issue of limited generalization caused by insufficient modeling of forgery representations, this paper proposes a channel-aware local–global representation learning network for generalizable deepfake detection. Specifically, we introduce a Local–Global Integration Vision Transformer (LGI-ViT) block that learns local and global representations and further integrates them with input features to capture more generalizable forgery cues at both fine-grained and global levels. Local representation learning is enhanced through coordinate convolution, while a hybrid convolution–Transformer architecture is employed to model global dependencies. Based on these representations, a residual connection is incorporated to integrate the combined local–global representation with the original input features. In addition, a Lightweight Channel Attention Network (LCAN) block is designed to strengthen interactions among feature channels and improve the discriminability of forgery-related representations. Experimental results demonstrate that the proposed network, trained on the FaceForensics++ (FF++) dataset, achieves cross-dataset AUC scores of 73.95% and 79.42% on the DeepFake Detection Challenge (DFDC) and Celeb-DF datasets, respectively. It outperforms the best-performing baseline among 11 competing models for generalizable detection by 1.01 percentage points on average, thereby validating its effectiveness in generalizable deepfake detection.
Förderung
- Wuxi Science and Technology Innovation Fund “Taihu Light” Science and Technology Tackling Program
- the Jiangsu Provincial Higher Vocational Colleges Engineering Technology Research and Development Center
- the Jiangsu Provincial University Excellent Scientific and Technological Innovation Team
Zitieren
APA
Kuang, L., Chen, B., & Shi, P. (2026). Channel-Aware Local–Global Representation Learning for Generalizable Deepfake Detection. Mathematics, 14(11), 1913. https://doi.org/10.3390/math14111913
Vancouver
Kuang L, Chen B, Shi P. Channel-Aware Local–Global Representation Learning for Generalizable Deepfake Detection. Mathematics. 2026;14(11):1913. doi:10.3390/math14111913
BibTeX
@article{Kuang2026,
author = {Kuang, Liang and Chen, Beijing and Shi, Pei},
title = {Channel-Aware Local–Global Representation Learning for Generalizable Deepfake Detection},
journal = {Mathematics},
year = {2026},
volume = {14},
number = {11},
pages = {1913},
doi = {10.3390/math14111913}
}
EUVIMED