Integration of Deep Learning and Traditional Artistic Styles for Automated Generation of Intangible Cultural Heritage Content
Abstract
This archive contains the complete implementation of the StyleGAN-Assisted Diffusion Model proposed in the associated manuscript "Integration of Deep Learning and Traditional Artistic Styles for Automated Generation of Intangible Cultural Heritage Content". The framework uses a two-stage cascaded strategy for category-conditioned intangible cultural heritage (ICH) image generation. Stage I trains a category-conditioned StyleGAN as a structural-prior generator that captures global layout, silhouette organisation and approximate motif placement. Stage II trains a latent diffusion backbone that refines local texture under latent-space structural injection and ICH-aware regional control, using deterministic edge-texture-category region parsing, softened masks, fixed category-specific cue templates, and a cultural-structure regularization term covering ink-painting composition, paper-cutting bilateral symmetry and ceramic motif repetition. The archive includes: the full source code for both training stages; category-conditioned latent inversion and alignment-cache construction; the ICH-aware regional partition and cue-assignment pipeline; all loss terms including the cultural-structure regularizers; the free-generation and reference-aligned evaluation protocols; all ablation, sensitivity, efficiency, bias and fine-detail diagnostic scripts; configuration files; prompt templates; split-generation scripts; and the fixed random seeds used in the paper. Three primary ICH categories…
Read the paper
Related studies
- Learning from First-Generation Qualitative Approaches in the IS Discipline: An Evolutionary View and Some Implications for Authors and Evaluators (PART 1/2)
- Commentary on the clinical and preclinical dosage limits of interstitially administered magnetic fluids for therapeutic hyperthermia based on current practice and efficacy models
- Prosthesis: New Technological Opportunities and Innovative Biomedical Devices
- The history and future of neural modeling for cochlear implants
- Ryanodine receptors: waking up from refractoriness
- Novel Transgenic Rabbit Model Sheds Light on the Puzzling Role of Matrix Metalloproteinase-12 in Atherosclerosis
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.