Imagen de Portada

Hierarchical symmetric GAN for Thangka image generation

Guardado en:
Publicado en:Heritage Science vol. 13, no. 1 (Dec 2025), p. 568
Autor principal: Hu, Wenjin
Otros Autores: Zhao, Yan, Yin, Lemei, Zhang, Guoquan
Publicado:
Springer Nature B.V.
Materias:
Feature extraction
Diffusion models
Art
Digital preservation
Datasets
Image processing
Discriminators
Optimization techniques
Generative artificial intelligence
Neural networks
Deities
Semantics
Cultural heritage
Acceso en línea:Citation/Abstract
Full Text
Full Text - PDF
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Resumen:Text-to-Thangka generation requires preserving both semantic accuracy and textural details. Current methods struggle with fine-grained feature extraction, multi-level feature integration, and discriminator overfitting due to limited Thangka data. We present HST-GAN, a novel framework combining parallel hybrid attention with differentiable symmetric augmentation. The architecture features a Parallel Spatial-Channel Attention module (PSCA) for precise localization of deity facial features and ritual object textures, along with a Hierarchical Feature Fusion Network (HLFN) for multi-scale alignment. The framework’s Differentiable Symmetric Augmentation (DiffAugment) dynamically adjusts discriminator inputs to prevent overfitting while improving generalization. On the T2IThangka dataset, HST-GAN achieves an Inception Score of 2.08 and reduces Fréchet Inception Distance to 87.91, demonstrating superior performance over baselines on the Oxford-102 benchmark.
ISSN:2050-7445
DOI:10.1038/s40494-025-02100-3
Fuente:Materials Science Database