Detecting and Tracking Sports Players with Random Forests and Context-Conditioned Motion Models

Player movements in team sports are often complex and highly correlated with both nearby and distant players. A single motion model would require many degrees of freedom to represent the full motion diversity of each player and could be difficult to use in practice. Instead, we introduce a set of Game Context Features extracted from noisy detection data to describe the current state of the match, such as how the players are spatially distributed. Our assumption is that players react to the current game situation in only a finite number of ways. As a result, we are able to select an appropriate simplified motion model for each player and at each time instant using a random decision forest which examines characteristics of individual trajectories and broad game context features derived from all current trajectories. Our context-conditioned motion models implicitly incorporate complex inter-object correlations while remaining tractable. We demonstrate significant performance improvements over existing multi-target tracking algorithms on basketball and field hockey sequences of several minutes in duration containing ten and twenty players respectively.