TY - JOUR
T1 - Promise Into Practice
T2 - Application of Computer Vision in Empirical Research on Social Distancing
AU - Bernasco, Wim
AU - Hoeben, Evelien
AU - Koelman, Dennis
AU - Liebst, Lasse Suonperä
AU - Thomas, Josephine
AU - Appelman, Joska
AU - Snoek, Cees
AU - Lindegaard, Marie Rosenkrantz
PY - 2023
Y1 - 2023
N2 - Social scientists increasingly use video data, but large-scale analysis of its content is often constrained by scarce manual coding resources. Upscaling may be possible with the application of automated coding procedures, which are being developed in the field of computer vision. Here, we introduce computer vision to social scientists, review the state-of-the-art in relevant subfields, and provide a working example of how computer vision can be applied in empirical sociological work. Our application involves defining a ground truth by human coders, developing an algorithm for automated coding, testing the performance of the algorithm against the ground truth, and running the algorithm on a large-scale dataset of CCTV images. The working example concerns monitoring social distancing behavior in public space over more than a year of the COVID-19 pandemic. Finally, we discuss prospects for the use of computer vision in empirical social science research and address technical and ethical challenges.
AB - Social scientists increasingly use video data, but large-scale analysis of its content is often constrained by scarce manual coding resources. Upscaling may be possible with the application of automated coding procedures, which are being developed in the field of computer vision. Here, we introduce computer vision to social scientists, review the state-of-the-art in relevant subfields, and provide a working example of how computer vision can be applied in empirical sociological work. Our application involves defining a ground truth by human coders, developing an algorithm for automated coding, testing the performance of the algorithm against the ground truth, and running the algorithm on a large-scale dataset of CCTV images. The working example concerns monitoring social distancing behavior in public space over more than a year of the COVID-19 pandemic. Finally, we discuss prospects for the use of computer vision in empirical social science research and address technical and ethical challenges.
KW - Faculty of Social Sciences
KW - Computer vision
KW - video data analysis
KW - deep learning
KW - pedestrian detection
KW - social distancing
U2 - 10.1177/00491241221099554
DO - 10.1177/00491241221099554
M3 - Journal article
VL - 52
SP - 1239
EP - 1287
JO - Sociological Methods and Research
JF - Sociological Methods and Research
SN - 0049-1241
IS - 3
ER -