Theses Master's

Computer Vision for Ethnographic Research

Balakrishnan, Kirthi

This thesis proposes a novel computer vision method for conducting large-scale and automated analysis of the commercial landscape of cities over time. The method employs ML techniques and leverages street view imagery to extract and analyze visual information, such as colors and languages used across storefronts, to identify changes in the characteristics of ethnic enclaves in cities.

The paper argues that this method could be used in the context of ethnographic research to study the cultural practices and beliefs of communities in cities like Detroit, New York, and Los Angeles, and to develop a digital tool that maps the quantifiable changes in the characteristics of ethnic enclaves over time. By using qualitative data to draw causal inferences between these changes and events that may have had an impact, the tool could be used by urban planners and sociologists to better understand and predict urban change in the context of ethnic enclaves and cities. The paper also proposes a framework for incorporating this data into urban planning and sociological research, providing a basis for further study in this area.

Files

  • thumnail for BalakrishnanKirthi_GSAPPUP_2023_Thesis.pdf BalakrishnanKirthi_GSAPPUP_2023_Thesis.pdf application/pdf 2.63 MB Download File

More About This Work

Academic Units
Urban Planning
Thesis Advisors
Vanky, Anthony P.
Degree
M.S., Columbia University
Published Here
August 16, 2023