Research on Fashion Contextual Style Recognition Methods Based on Scene Graph Technology
Journal of Design Intelligence · 24.06.2026 · 1 (3) · 10-24
DOI: 10.64509/jdi.13.115 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Design Intelligence, online erschienen am 24.06.2026. Verfasst von Yuming Mai und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Driven by the intelligent and personalized transformation of the garment industry, fashion contextual style recognition has emerged as a pivotal technology bridging visual understanding and business decision-making. However, traditional methods anchor styles to isolated garment attributes, failing to capture the deep semantics of entity interactions in complex visual scenes. To address these limitations, we propose a fashion contextual style recognition method based on unbiased Scene Graph Generation (SGG) and hierarchical context fusion. We developed a hierarchical context-aware network that deconstructs fashion contexts into three dimensions: entity context, global context, and scene context, integrated through graph embedding. Furthermore, a Causal Intervention mechanism based on Total Direct Effect (TDE) is introduced to mitigate the long-tail bias inherent in unconstrained scenes. Experimental results on the Visual Genome dataset demonstrate that our approach significantly outperforms baseline models in recognition accuracy and robustness. This research extends the definition of fashion style from visual attributes to holistic situational semantics, providing a new perspective for intelligent fashion analysis.
Zitieren
APA
Mai, Y., Zhao, X., & Hong, Y. (2026). Research on Fashion Contextual Style Recognition Methods Based on Scene Graph Technology. Journal of Design Intelligence, 1(3), 10-24. https://doi.org/10.64509/jdi.13.115
Vancouver
Mai Y, Zhao X, Hong Y. Research on Fashion Contextual Style Recognition Methods Based on Scene Graph Technology. Journal of Design Intelligence. 2026;1(3):10-24. doi:10.64509/jdi.13.115
BibTeX
@article{Mai2026,
author = {Mai, Yuming and Zhao, Xin and Hong, Yan},
title = {Research on Fashion Contextual Style Recognition Methods Based on Scene Graph Technology},
journal = {Journal of Design Intelligence},
year = {2026},
volume = {1},
number = {3},
pages = {10--24},
doi = {10.64509/jdi.13.115}
}
EUVIMED