Soccerwatch

One of the founders of soccerwatch.tv is the father of two football-mad children but often could not make it to their matches. This gave him the idea of installing a webcam at the football pitch, so he could follow the matches live. A prototype was made from a plastic box bought at a DIY store with a hole cut out of it and a camera installed inside, and the match was then uploaded to YouTube.

soccerwatch.tv uses digital camera technology to record amateur football matches and broadcast them live, on demand or as highlights. According to the German Football Association, 1.8 million football matches take place in Germany every year, of which 99 per cent of them are not recorded. The reason? Previous solutions are cumbersome, expensive and substandard.

It is critical for every AI project that everyone involved develop a mutual understanding of the business idea, the existing IT structure, the structure and quality of the data and what information the data contains. To do so, a requirements analysis was carried out with soccerwatch.tv and adesso experts as part of a workshop in the Interaction Room. Then a suitable implementation plan was created and appropriate architecture and technology were selected.

Open source software combined with a cloud solution proved to be the perfect choice for soccerwatch.tv. On the one hand, the start-up did not have the finances or staff needed to start with their own computer cluster immediately, but on the other hand, a cloud system is exceptionally well-suited for their business idea. The company can scale the capacity up or down based on demand with every additional camera system installed.

The experts collected the video footage from a few camera systems and used it to develop the algorithm as a first step. The goal was to prove the functionality and feasibility of the concept developed in the workshop. The development was entirely agile. Agility in regard to AI projects has the critical advantage of being able to bring in analysts at a very early stage in the modelling process. In this manner, those responsible can recognise early on whether the implementation delivers the desired benefit, unlike traditional approaches where the data for analysts is only available right at the end.

The parties involved integrated the algorithm into the cloud solution after they had successfully proven the viability of the approach. The two essential requirements: scalability and real-time capability. Weekend after weekend, 10,000 amateur football matches take place at the same time in Germany. Every camera system installed increases the processing load that the system needs to process in parallel in order to stream all these matches live.

soccerwatch.tv currently has approximately XX camera systems installed. The highlight of the still developing history: The streaming start-up and the German Football Association have signed a partnership. Amateur coaches and players can use soccerwatch.tv Analytics to have access to stats such as running activity, one-on-one battle ratios and tracking values, just like those of the professionals.