Dual-Space Hierarchical Learning for Deepfake Detection
IEEE Signal Processing Letters · 2026 · 33 · 2280-2284
DOI: 10.1109/lsp.2026.3696217 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Signal Processing Letters, 2026. Verfasst von Shanshan Han und 4 weiteren Autor:innen. Die Arbeit nennt 38 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Science and Technology Research Program of Chongqing Municipal Education.
Abstract (Originalsprache)
Most existing deepfake detection methods learn visual artifacts in Euclidean space, overlooking the intrinsic hierarchical relationships among forgery samples. In practice, manipulated images deviate progressively from real ones, forming a hierarchy that Euclidean geometry struggles to model, while hyperbolic space better captures such structures due to its exponential capacity. Motivated by this observation, we propose a dual-space hierarchical learning framework that jointly models Euclidean visual semantics and hyperbolic hierarchical representations. Specifically, Euclidean features are projected into hyperbolic space, where a Hyperbolic Hierarchy-Aware Attention (HHAA) module is introduced to capture hierarchical dependencies among samples. To effectively integrate complementary information from the two spaces, a Dual-Space Gated Fusion (DSGF) module is designed to adaptively inject hierarchical cues into Euclidean features. A joint supervision strategy is further applied to enhance discriminative representation learning. Extensive experiments on multiple deepfake detection benchmarks demonstrate that the proposed method improves detection performance and generalization ability.
Förderung
- National Natural Science Foundation of China
- Science and Technology Research Program of Chongqing Municipal Education
Zitieren
APA
Han, S., Yuan, Y., Wang, H., Yuan, L., & Gao, X. (2026). Dual-Space Hierarchical Learning for Deepfake Detection. IEEE Signal Processing Letters, 33, 2280-2284. https://doi.org/10.1109/lsp.2026.3696217
Vancouver
Han S, Yuan Y, Wang H, Yuan L, Gao X. Dual-Space Hierarchical Learning for Deepfake Detection. IEEE Signal Processing Letters. 2026;33:2280-2284. doi:10.1109/lsp.2026.3696217
BibTeX
@article{Han2026,
author = {Han, Shanshan and Yuan, Yifan and Wang, Hongyi and Yuan, Lin and Gao, Xinbo},
title = {Dual-Space Hierarchical Learning for Deepfake Detection},
journal = {IEEE Signal Processing Letters},
year = {2026},
volume = {33},
pages = {2280--2284},
doi = {10.1109/lsp.2026.3696217}
}
EUVIMED