AI in Cycling: an introduction to the basics of machine learning and
computer vision for analysing training and race data

On 1 July 2025 a masterclass will be given by  IDLab Sports Data Science team of Ghent University
Location: Lilliad Learning Center

Coffee breaks and lunch included

Time 09.00-17.00

Introduction

Data analysis in cycling is revolutionizing the sport. More and more, teams use data science
to evaluate rider performance and identify strengths and weaknesses of their riders. This
information aids in recruitment and the development of targeted training programs.
Additionally, data analysis provides insights into competitor strategies, allowing coaches to
make informed tactical decisions. Beyond performance optimization, data also plays a
crucial role in enhancing the fan experience through storytelling and ensuring the safety of
riders and their fans.

The masterclass focuses on each of these topics: performance, safety and storytelling. 

The masterclass will be a 1-day intensive course bringing together junior and senior
academics and practitioners/professionals active within the cycling sector who want to learn
the basics of machine learning and computer vision for analysing training and race data. Main focus will be on
sports scientists who want to gather more AI/data science skills, but also
experienced data scientists who want to apply their skills on real-world cycling data (that is
collected in cycling projects over the last years) are welcome.

Program

During the morning a general introduction and explanation of the basic concepts of AI
(cleaning, balancing, clustering, classification, …) and computer vision (object detection,
segmentation, pose estimation and video analysis) will be given. There will be a demonstration based on cycling
data that are collected on the track and on the road. Python notebooks (in
Google Colab) with the most important scripts will be provided. In the afternoon, participants can use these
exemplary scripts in a mini hackathon in which they will work in groups to create a
demonstrator for one of the cycling use cases for which we will provide them a multimodal
dataset with a mix of sensor and video data. We will close the day with a presentation of the
hackathon results.

About the instructor

IDLab Sports Data Science team of Ghent University has built up proven AI and computer vision expertise in sports data
science over the last 10 years, with the majority of their projects in cycling, in collaboration with federations, the
UCI and the broadcasters. Those projects and the tools that IDLab has investigated and
implemented for them, will form the basis of this masterclass.

REGISTRATION FEES MASTERCLASSES SCIENCE & CYCLING 2025

(Including 21% French vat)