Semantic-Driven Deep Learning Network for Visual Communication in Compound Chinese Character Design
International Journal on Semantic Web and Information Systems · 01.04.2026 · 22 (1) · 1-22
DOI: 10.4018/ijswis.406077 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal on Semantic Web and Information Systems, erschienen am 01.04.2026. Verfasst von Wei Zhu. Die Arbeit nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 3.0).
Abstract (Originalsprache)
Synthesizing compound Chinese characters is pivotal for modern typography and interaction design, yet intricate stroke coupling often leads to structural collapse in existing models. Current methods, which rely on loose implicit mappings, frequently produce disordered layouts and blurred artifacts. To rectify this, the author proposes latent structure-guided diffusion, which explicitly injects geometric priors into the diffusion process. By coordinating a latent cross-condition projector for semantic alignment and a latent deformable layout field for differentiable stroke manipulation, latent structure-guided diffusion ensures precise compositional control. Experiments confirmed its superior performance, noting that it achieved a structural similarity index of 0.95 and a mean edge gradient of 0.94, thus significantly outperforming state-of-the-art baselines in structural fidelity and stylistic coherence.
Zitieren
APA
Zhu, W. (2026). Semantic-Driven Deep Learning Network for Visual Communication in Compound Chinese Character Design. International Journal on Semantic Web and Information Systems, 22(1), 1-22. https://doi.org/10.4018/ijswis.406077
Vancouver
Zhu W. Semantic-Driven Deep Learning Network for Visual Communication in Compound Chinese Character Design. International Journal on Semantic Web and Information Systems. 2026;22(1):1-22. doi:10.4018/ijswis.406077
BibTeX
@article{Zhu2026,
author = {Zhu, Wei},
title = {Semantic-Driven Deep Learning Network for Visual Communication in Compound Chinese Character Design},
journal = {International Journal on Semantic Web and Information Systems},
year = {2026},
volume = {22},
number = {1},
pages = {1--22},
doi = {10.4018/ijswis.406077}
}
EUVIMED