Understanding global pedestrian behaviour in 4965 cities with dashcam videos on YouTube

Alam, Md. S., Martens, M. H., Bazilinska, O., Bazilinskyy, P.

Submitted for publication.

ABSTRACT The interactions between cars and pedestrians should be designed to be understandable and safe worldwide. Although previous research has studied vehicle-pedestrian interactions within specific cities or countries, this study offers a more scalable and robust approach by examining pedestrian behaviour worldwide. We present a dataset (CROWD), which includes 12260 hours of YouTube day and night dashcam footage from 4965 cities in 238 countries and territories. The included videos feature continuous urban driving, are at least 300 s long, feature no atypical events and represent everyday conditions, and are from cities with a minimum of 10,000 population. We detected pedestrian movements, focussing on the speed and the pedestrian crossing decision time during road crossings based on the bounding boxes given by YOLOv11x. The results revealed statistically significant variations in pedestrian behaviour influenced by socioeconomic and environmental factors such as Gross Metropolitan Product, traffic-related mortality, Gini coefficient, traffic index, average age and literacy.