[AI-Generated Summary]: This comprehensive article by Jia Li and Wang Yanling from Sichuan University examines the current state and future prospects of art image database construction from a digital humanities perspective. The authors categorize four types of art image databases: Digital Photo Archives (like Rembrandt Database), Iconographic Databases (such as Warburg Institute’s database), Digital Libraries (exemplified by Harvard Digital Library), and Online Galleries (including museum collections and Google Arts & Culture). The paper explores the methodological shift from “distant reading” in literature to “distant viewing” in art history, discussing how computational methods enable analysis of large visual corpora to reveal patterns invisible to traditional approaches. The authors emphasize that while quantitative analysis tools are valuable, databases should serve as comprehensive research infrastructures combining multiple digital humanities methods. They advocate for improved data interconnection, standardized metadata frameworks like IIIF, and enhanced digital literacy among researchers to fully leverage these resources for innovative art historical research.
Language: zh