Multi-perspective perception of city image in Shanghai via massive short videos

• Innovative framework analyzes city image using short videos and deep learning. • Multi-modal data integration provides a holistic interpretation of urban environments. • Core urban areas are concentrated while peripheral areas show dispersed spatial patterns. • Public spaces’ visual and cultural traits enhance city recognition and appeal. The concept of city image plays a pivotal role in urban branding, serving as a strategic tool for promoting sustainable urban development. A city’s image not only shapes its identity but also enhances its global competitiveness, attracting tourism, investment, and new residents. Traditional media such as newspapers and television offer limited perspectives on city image, whereas emerging platforms like short video apps provide richer and more diverse insights. Nevertheless, existing studies remain dominated by traditional media analysis, with insufficient examination of how short video platforms represent the spatial, landscape, social, and emotional dimensions of a city’s image. This study proposed a novel approach by leveraging the diverse and extensive short video data from Douyin to thoroughly deconstruct Shanghai’s comprehensive city image. We constructed a geotagged short video database, categorizing landscape and social images based on visual features and assessing emotional tones through audio-text analysis. The findings revealed distinct spatial patterns, with core areas highly concentrated and peripheral areas more dispersed. The analysis further identified nine core image areas, three major development axes, and ten prominent zones that defined Shanghai’s unique geographic and cultural identity. This approach could provide a quantitative characterization of city image, highlighting the potential for improved public space design and enhanced citizen engagement through short videos. Moreover, these results underscored the significant role of social media platforms, especially short video apps, in shaping public perception of cities. Such insights could also offer new perspectives for city image research and valuable guidance for urban planning and branding strategies.

Authors

Institutions

Publication Details

Journal
International Journal of Applied Earth Observation and Geoinformation
Published
2025-09-25
DOI
https://doi.org/10.1016/j.jag.2025.104844
Citations
1
Primary Topic
Remote Sensing and Land Use
Type
article
Field-Weighted Citation Impact
1.30
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Multi-perspective perception of city image in Shanghai via massive short videos

Kai Cao, Minxin Chen, Zhen Wei
1 citations
International Journal of Applied Earth Observation and Geoinformation
Remote Sensing and Land Use
1.30
article

Multi-perspective perception of city image in Shanghai via massive short videos

Kai Cao, Minxin Chen, Zhen Wei
article en
1 citations

Abstract

• Innovative framework analyzes city image using short videos and deep learning. • Multi-modal data integration provides a holistic interpretation of urban environments. • Core urban areas are concentrated while peripheral areas show dispersed spatial patterns. • Public spaces’ visual and cultural traits enhance city recognition and appeal. The concept of city image plays a pivotal role in urban branding, serving as a strategic tool for promoting sustainable urban development. A city’s image not only shapes its identity but also enhances its global competitiveness, attracting tourism, investment, and new residents. Traditional media such as newspapers and television offer limited perspectives on city image, whereas emerging platforms like short video apps provide richer and more diverse insights. Nevertheless, existing studies remain dominated by traditional media analysis, with insufficient examination of how short video platforms represent the spatial, landscape, social, and emotional dimensions of a city’s image. This study proposed a novel approach by leveraging the diverse and extensive short video data from Douyin to thoroughly deconstruct Shanghai’s comprehensive city image. We constructed a geotagged short video database, categorizing landscape and social images based on visual features and assessing emotional tones through audio-text analysis. The findings revealed distinct spatial patterns, with core areas highly concentrated and peripheral areas more dispersed. The analysis further identified nine core image areas, three major development axes, and ten prominent zones that defined Shanghai’s unique geographic and cultural identity. This approach could provide a quantitative characterization of city image, highlighting the potential for improved public space design and enhanced citizen engagement through short videos. Moreover, these results underscored the significant role of social media platforms, especially short video apps, in shaping public perception of cities. Such insights could also offer new perspectives for city image research and valuable guidance for urban planning and branding strategies.

International Journal of Applied Earth Observation and GeoinformationVol. 144
Peking University (CN), East China Normal University (CN)
Openalex Percentile: Top 18%
Remote Sensing and Land Use
1.30
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.