Learning patterns of tourist movement and photography from geotagged photos at archaeological heritage sites in Cuzco, Peru. (February 2021)

• Record Type: Journal Article
• Authors: Payntar, Nicole D.; Hsiao, Wei-Lin; Covey, R. Alan; Grauman, Kristen
• Is Part Of: Tourism management. Volume 82(2021)
• Publication Date: 2021-02
• Subjects: Geotagged photos -- Anthropology and computer vision -- Clustering -- Visuality -- Heritage tourism -- Visual recognition -- Peru; Tourism -- Periodicals; 338.4791
• DOI: 10.1016/j.tourman.2020.104165
• Languages: English
• ISSNs: 0261-5177

Abstract

Abstract: The popularity of media sharing platforms in recent decades has provided an abundance of open source data that remains underutilized by heritage scholars. By pairing geotagged internet photographs with machine learning and computer vision algorithms, we build upon the current theoretical discourse of anthropology associated with visuality and heritage tourism to identify travel patterns across a known archaeological heritage circuit, and quantify visual culture and experiences in Cuzco, Peru. Leveraging large-scale in-the-wild tourist photos, our goals are to (1) understand how the intensification of tourism intersects with heritage regulations and social media, aiding in the articulation of travel patterns across Cuzco's heritage landscape; and to (2) assess how aesthetic preferences and visuality become entangled with the rapidly evolving expectations of tourists, whose travel narratives are curated on social media and grounded in historic site representations. Highlights: First study to quantify visual culture of heritage tourism in Cuzco, Peru. Tourist movements are attributable to changes in available BTC ticket packages. Homogeneity of image clusters high at large heritage sites, diverse at small sites. Common image themes: ruins, mountains, terracing, indigenous peoples, markets. Affinity propagation, t-sne, scene labeling as tools for future heritage studies.