In past decades, great attention has been paid to organization and integration in public digital cultural resources, but now the main work transfers from storage to utilization, focusing on user experience and cultural service. Taking Tang Tri-color as the object, based on user's portrait and Grounded Theory, this study builds up specific personal requirements and encodes their different semantic requirements on the display of Tang Tri-color. Then it constructs a semantic description model, showing the core need of semantic description on digital cultural resources from all aspects including identification description, basic description, value description and extended description. Relying on the preceding steps, the metadata framework on display of Tang Tri-color resources is established.