Carlo Ratti, Enrico Costanza, and Justin Moe (in collaboration with Matteo Motterlini)

Information about the movement of soccer players on the field during a match can be useful for strategic and physiologic analysis, directed at improving the performance of the players and that of the team. The use of electronic sensing techniques – mostly computer vision – for automatically tracking the players has been an area of active research for many years. An optimum solution, however, has yet to be found. The problem is challenging from a technical point of view because the sensing area is large compared to the moving actors (players), the actors move fast and occlusion and congestion occur frequently. This research project, in cooperation with the Italian soccer team A.C. Milan, aims to identify and develop new real-time tracking systems and to interpret from a strategic point of view the data they provide.