Efficient searching and representation of tracking data in healthcare and sports

The task of finding decisive – or otherwise interesting – scenes in the tracking data recorded during games such as soccer or ice hockey, evaluating them, and presenting potential solutions for situations that arise during the game, is a fundamental task for game analysis in sports clubs and associations, and for media coverage of sporting events. The current standard procedure involves manual entry of appropriate labels during a game, or time-consuming searches of video footage after the event.

The goal of this interdisciplinary application is to develop a search engine for sports scenes that can use position data to find similar situations in databases of past games. In addition to finding game situations and configurations, a subsequent step is evaluating them in terms of execution. In a final step, the system should be able to propose solutions for a given situation (optimization).

This automated process is intended for games in which tracking data are already widely available (such as soccer, basketball, or football), or will be in the near future (such as hockey). For clubs and associations, tracking data are the only machine-collected data for objective game analysis. Using these data, the project aims to help users achieve more effective training management and talent scouting, for instance by automatically identifying weaknesses, or collating scenes which a given player resolved either successfully or unsuccessfully.