Welcome to Pedestrian Tracking’s documentation!

Contents:

• Quickstart for Pedestrian Tracking
  • Basic Motion Detection
  • Fixed Background Subtraction
  • Mixture of Gaussians Background Subtraction
  • Basic Usage
  • Results
  • Customisation
• Guidelines for Pedestrian Tracking Video
  • Vantage Point
  • Entrances and Exits
  • Video Format
  • ffmpeg
• Optimising the Pedestrian Tracker for Your Videos
  • Erosion and Dilation
  • Blob Parameters
  • Using Grid View
• Validating and Visualising Pedestrian Trails
  • Validation
  • Drawing Modes
• Overview of Files and Tools

These resources are available to analyse and visualise pedestrian traffic using computer vision. These files can generate behavioural maps and statistics about passers-by such as walking speed and direction.

If using this code or the associated materials, please cite this source:

Williamson, J.R. and Williamson, J. Analysing Pedestrian Traffic Around Public Displays. In the Proceedings of Pervasive Displays 2014. ACM, New York, USA.

The original publication is available at: Performative Interaction Blog and The ACM Digital Library.

All the source code and data is available at: GitHub, or download the zip.

This documentation is a work in progress! If you have specific questions, you can direct them to Julie.Williamson@glasgow.ac.uk.

Dependencies

All resources are developed in Python (2.7). Video preparation was completed using ffmpeg (mjpeg codec used, although any format supported by OpenCV is acceptable).

Please ensure the following dependencies are installed before using this tool:

• OpenCV
• Numpy
• PyLab
• SciPy
• Munkres Hungarian Algorithm

Indices and tables

• Index
• Module Index
• Search Page