Scholarlink

<p>Pedestrian trajectories are used to learn about human behavior in public space and the impact of spatial features on pedestrian flows. Currently, these trajectories are collected manually, with self-tracking devices, or with video cameras. Even when trajectories are obtained using computational techniques, such as using computer vision to trace them in space, these datasets are not made available for reproducibility or comparative studies between different locations. To close this gap, this paper makes available the data of pedestrian trajectories collected in 39 European squares. Firstly, we summarize the data collection process which was based on collecting footage from publicly available webcams. Secondly, we describe the process of trajectory extraction entailing object detection, tracking, and georeferencing. Lastly, we describe the data cleaning and validation steps that lead to the final dataset. The dataset ultimately includes 348,300 pedestrian trajectories extracted from 193 hours of video footage, collected at different times of the day, during working days and weekends, and during the Spring and Summer season.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Pedestrian Trajectory Dataset of Public European Squares

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月10日

查看全文
  • Nils Wolff,
  • Layne Perry,
  • Titus Venverloo,
  • Geertje Slingerland,
  • Jessica Wreyford,
  • Paolo Santi,
  • Fábio Duarte

摘要

Pedestrian trajectories are used to learn about human behavior in public space and the impact of spatial features on pedestrian flows. Currently, these trajectories are collected manually, with self-tracking devices, or with video cameras. Even when trajectories are obtained using computational techniques, such as using computer vision to trace them in space, these datasets are not made available for reproducibility or comparative studies between different locations. To close this gap, this paper makes available the data of pedestrian trajectories collected in 39 European squares. Firstly, we summarize the data collection process which was based on collecting footage from publicly available webcams. Secondly, we describe the process of trajectory extraction entailing object detection, tracking, and georeferencing. Lastly, we describe the data cleaning and validation steps that lead to the final dataset. The dataset ultimately includes 348,300 pedestrian trajectories extracted from 193 hours of video footage, collected at different times of the day, during working days and weekends, and during the Spring and Summer season.