Research on the Construction of a Multimodal Cognitive Graph of Textile and Clothing

With the development of the textile and clothing industry, understanding and grasping the characteristics and properties of textiles and clothing has become increasingly important. This article proposes a multimodal textile and clothing cognition graph(CG) construction method to comprehend textiles and clothing better. This method combines knowledge graph(KG) and multimodal data processing techniques to integrate various information about textiles and clothing into a unified CG. Firstly, critical information is extracted from textual and image data related to textiles and clothing using web crawling techniques. Then, utilizing modeling and representation methods of KG, the extracted information is organized into entities and relationships, thus constructing a CG for textiles and clothing. Finally, browsing and analysis of the textile and clothing CG are realized through graph visualization and interactive technologies. Experimental results demonstrate that this method can effectively construct a multimodal CG for textiles and clothing, supporting research and applications in the textile and clothing field.