Peloton tracking and analysis from the 2016 Tour de France
The work associated with this project is described in a manuscript entitled "How vision governs the collective behavior of cycling pelotons" by Belden et al., along with an electronic supplementary material document. We investigate properties of densely packed groups of bicycle racers, which are known as cycling pelotons. These pelotons exhibit features of collective animal behavior, including emergent behavior from inter-individual interactions. In this data set, we classify global shapes of the peloton, and identify and track individual cyclists to determine the details of network structure. We also investigate motion waves that propagate through the pelotons.
.MAT, .M, .FIG, .XLSX, .TIFF, .EPS, .TXT, .PNG
MatLab or open source program required to open .mat files.
Office of Naval Research
Utah State University
Office of Naval Research N00014-19-1-2059
The raw data come from video footage of the 2016 Tour de France professional bicycle stage race, which is filmed from a helicopter to provide an overhead view of the pelotons. Video clips of interested are extracted and associated with certain metadata about the details of the race at the instant of the clip. Video clips are parsed into individual images at the time spacing corresponding to the frame rate of capture. A set of image processing algorithms is used to extract rider locations, network structure, and wave propagation behavior in a metric reference frame; these algorithms are described in "How vision governs the collective behavior of cycling pelotons" by Belden et al., and in the supplementary material document supplied with the paper.
See attached README file
This work is licensed under a Creative Commons Attribution 4.0 License.
Truscott, Tadd T. and Belden, Jesse, "Peloton tracking and analysis from the 2016 Tour de France" (2019). Browse all Datasets. Paper 70.
FilesREADME.txt (4 kB)