Development and evaluation of algorithms for determining kite characteristics using 360° video data in kitesurfing
Project duration: 01/2026 – 12/2026
Kitesurfing is an extremely dynamic sport. In competitions and training sessions, athletes reach speeds of up to 60 km/h with their equipment. A conventional technical analysis, such as using video recordings or a visual assessment from a motorboat travelling alongside, is practically impossible under these conditions. Athletes are therefore currently being equipped with 360° cameras attached to their helmets. Based on the video footage captured in this way, technical analysis is carried out after the training session by manually assessing the height and shape of the kite using individual, selected still images. This process is very time-consuming and, moreover, never captures the entire training session.
As part of this innovation project, funded by the BISp, machine learning (ML) techniques are therefore being used to develop algorithms that enable the automated extraction of technical parameters relevant to kitesurfing from video footage. The approach involves, amongst other things, adapting segmentation and object detection methods to identify the kite, the lines and the horizon. The aim is to significantly improve the analysis of kitesurfing by quantifying the handling of the kite throughout the entire training session.
Contact
Prof. Dr.-Ing. Mirco Fuchs
Chair of Computer Vision and Machine Learning
Telephone: +49 (0)341 3076 3104
Email: mirco(dot)fuchs(at)htwk-leipzig.de
M.Sc. Daniel Matthes
Research Assistant
Telephone: +49 (0)341 3076 3208
Email: daniel(dot)matthes(at)htwk-leipzig.de
