Research on Computer Vision-Assisted Quantitative Analysis and Knowledge Graph Construction of Intangible Cultural Heritage Inheritors’ Skills

The tacit knowledge embedded in intangible cultural heritage skills is difficult to capture, quantify, and systematically preserve using conventional documentation approaches. To address this challenge, this study proposes a computer vision-assisted framework for quantitative analysis and knowledge graph construction of intangible cultural heritage inheritors’ skills. Human skeletal key-point detection, motion trajectory extraction, and force-related feature analysis are employed to achieve digital representation of operational behaviors. Ontology modeling and semantic association techniques are further integrated to construct a multimodal knowledge graph describing relationships among operational skills, cultural semantics, and inheritance knowledge. The proposed framework enables quantitative characterization of skill features and supports intelligent organization and retrieval of heritage knowledge. The study provides an effective solution for digital preservation and offers methodological references for computer vision, multimodal sensing, knowledge representation, and intelligent cultural-information systems.