Deep Learning for Automated Pre-Play Analysis in American Football Formations
The annotation and analysis of sporting videos is a tedious task. Once automated, it will be beneficial to players, coaches and spectators. This automation could be especially beneficial to American football as it is the most popular sport in the United States. The manual annotation and analysis on recorded videos of American Football games is a tedious and inefficient process. Most college football programs are currently focused on annotating the offensive formations in order to develop game plans. In order to advance research on this application, we used computer vision and deep-learning to analyze an image of a football game just before it began. This analysis involves locating football players, labeling their roles or positions, and identifying the offensive team’s formation.