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.



Document Type




File Format


Viewing Instructions

MatLab or open source program required to open .mat files.

Publication Date



Office of Naval Research


Utah State University

Award Number

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.

Referenced by

Belden, J., Mansoor, M. M., Hellum, A., Rahman, S. R., Meyer, A., Pease, C., Pacheco, J., Koziol, S., & Truscott, T. T. (2019). How vision governs the collective behaviour of dense cycling pelotons. Journal of The Royal Society Interface, 16(156), 20190197.



Code Lists

See attached README file


data supports forthcoming article and will be released upon article publication.


Mechanical Engineering


Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.



Additional Files

README.txt (4 kB)
MD5: 93d37a3622396b1b7c22b0e0074c0f56 (3147 kB)
MD5: 56bff06c6fed15162997045e8ff9457d (475243 kB)
MD5: de1597b21a436012e8080d99a058da88 (107720 kB)
MD5: f858438ea7d1df60fb8c4d35b02e89a2 (228494 kB)
MD5: 2d4245ac0cdd426b341ae4ede2c2cf42 (342369 kB)
MD5: 6549653ef8f61646c95f25f6a59d51aa (177165 kB)
MD5: 8fabcf80afad453e494bc6561fca1024 (177165 kB)
MD5: 8fabcf80afad453e494bc6561fca1024 (243521 kB)
MD5: dadf4df9961663615f50b750996fcd76 (224754 kB)
MD5: 0443bd6db7c756b86703c0e05698d5f8 (425524 kB)
MD5: d9a3f9b8e9f05d301929985afddf0f7a (299491 kB)
MD5: 6ddb726f1eed9ba04b820fe41874b4c3 (229301 kB)
MD5: 745f047bfbd2c7f108271bfd46e2d93a (112884 kB)
MD5: 2fd6c37723fe63e94d8eacbfee5fb695 (268187 kB)
MD5: 0ba548a0d045d436e861a3f16afd472f (175849 kB)
MD5: 097544d87ab2976325a68014b4e2fdd0 (140526 kB)
MD5: faf9a3330c74145d6920b0bfb8d77af3 (84379 kB)
MD5: 8ff353e747c8b80cf1d6226fac523c7e (227095 kB)
MD5: e8dc83cde424d3df145d5dfeaac5e697 (201883 kB)
MD5: d03965f16e24f3abcd48929d4f2b9e9f (191951 kB)
MD5: bf2bcddd8f77ba9bc2b8b34bb1461745 (114488 kB)
MD5: a02f2bd95034b2af3433a0a2a6998225 (1095479 kB)
MD5: cb3e39269043709abb48d0f132271017 (183876 kB)
MD5: a5e181cce1334ffaf4a97aa9ca7628af (659967 kB)
MD5: af85833c58dfefffbf50c9f6b1babb7a (238800 kB)
MD5: d2de4de9d56626e358e99d70cdc0d979