BR TV report · BISp project SurfD
AI-assisted motion analysis for coaching aerials in surfing
How can the execution of an aerial in surfing be analysed objectively and coaching be improved in a targeted way? A research project of the Movement Science unit is developing a digital assistant that uses an ordinary video to analyse the movement in each phase of the jump and delivers understandable feedback within a few minutes.
Using computer vision and pose estimation, body positions and joint points are captured in the video – without any additional sensors or markers. On this basis, a biomechanical model calculates movement parameters such as joint angles or the trajectory of the body’s centre of mass during take-off, the flight phase and landing.
Artificial intelligence then translates the biomechanical measurements into compact feedback on key technical elements, their strengths and the potential for improvement in motor learning and technique training. The goal: coaches should be able to support surfers in a targeted way, right at the water’s edge.
One principle applies: biomechanics, not AI, forms the basis of the analysis. The measurements provide the most objective basis possible, while AI helps to interpret them in an understandable way. The final assessment and training decisions remain with the coaches.
Funded by the Federal Institute of Sport Science (BISp) in cooperation with the German Surfing Association, O2 Surftown MUC and the Olympic Training Centre Bavaria.
TV report
Bavarian Broadcasting (BR) featured the project in “Abendschau – Der Süden”: Watch the report on br.de (in German)
