Answers are grounded in this paper and cite their sources.
Deep learning study demonstrates 96.83% accuracy in classifying southwest traditional crafts using an attention fusion network, highlighting viable tools for digital cultural archiving.
Southwest traditional crafts face recognition challenges due to intricate textures, variable scales, and a deficiency in specialized data. This study proposes the Scale-Wise Convolutional Refinement Attention Fusion Network (SW‑CraftNet), which integrates multi-scale features and dual attention mechanisms. Based on the public Chinese Intangible Cultural Heritage Image Dataset, a specialized refined dataset covering five major categories and twenty subcategories of southwest traditional crafts is constructed to supplement the domain-specific data resource. A multi-scale feature fusion module is designed to balance macro-composition and micro-stitchery. Furthermore, a channel-spatial attention mechanism is introduced to adaptively enhance critical cultural features. Experimental results indicate that the model achieves 96.83% accuracy on the Chinese Intangible Cultural Heritage Image Dataset. This performance outperforms mainstream architectures such as ResNet50, showing competitive performance in complex categories like Miao embroidery and lacquerware. The proposed model supports high-precision classification and retrieval of craft images, thereby providing a technical foundation for the digital archiving of intangible cultural heritage. These findings effectively promote the effective integration of technology and ethnic culture.
No takes yet. Share an insight, caveat, or question.
Liu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: