EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep learning architectures for multiclass clothing recognition as the semantic core of automated virtual try-on systems

Roman Chekhmestruk, Olena Voitsekhovska, Svitlana Kyrylashchuk

Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska · 30.06.2026 · 16 (2) · 162-172

DOI: 10.35784/iapgos.7957 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska, online erschienen am 30.06.2026. Verfasst von Roman Chekhmestruk und 2 weiteren Autor:innen. Die Arbeit nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This article examines and substantiates the choice of deep learning architectures for multiclass clothing classification integrated into virtual try-on (VTO) systems. Systematically compared ResNet-50, EfficientNet-B4, and Vision Transformer (ViT-B/16) on DeepFashion2 and ModaNet datasets. ViT-B/16 achieved the highest accuracy of 92.4% Top-1 on DeepFashion2 and 88.9% on ModaNet, demonstrating an average cross-dataset accuracy drop of 3.9 percentage points, the smallest among evaluated models. Preliminary U2-Net segmentation statistically significantly improved macro-F1 for all architectures (p < 0.001), with an average gain of 3.2 percentage points and reduction of the studio-to-street domain gap from 11 to 6 percentage points. EfficientNet-B4 provided the optimal accuracy-to-latency ratio, achieving 87% Top-1 accuracy at 60 FPS on consumer hardware (RTX 3060), while ViT-B/16 required optimization to maintain 45 FPS. The recommended strategy for industrial VTO systems combines U2-Net segmentation with architecture selection based on target platform capabilities, balancing visual fidelity and computational efficiency.

Zitieren

APA

Chekhmestruk, R., Voitsekhovska, O., & Kyrylashchuk, S. (2026). Deep learning architectures for multiclass clothing recognition as the semantic core of automated virtual try-on systems. Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska, 16(2), 162-172. https://doi.org/10.35784/iapgos.7957

Vancouver

Chekhmestruk R, Voitsekhovska O, Kyrylashchuk S. Deep learning architectures for multiclass clothing recognition as the semantic core of automated virtual try-on systems. Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska. 2026;16(2):162-172. doi:10.35784/iapgos.7957

BibTeX
@article{Chekhmestruk2026,
  author = {Chekhmestruk, Roman and Voitsekhovska, Olena and Kyrylashchuk, Svitlana},
  title = {Deep learning architectures for multiclass clothing recognition as the semantic core of automated virtual try-on systems},
  journal = {Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska},
  year = {2026},
  volume = {16},
  number = {2},
  pages = {162--172},
  doi = {10.35784/iapgos.7957}
}
Kostenlos starten →