Tracking Objects from Multiple Soccer Videos and Recognizing Events

This paper presents a novel way of recognizing events from a soccer match video sequence. After tracking players and the ball, events such as passing, kicking, having, scoring and struggling for the ball can be inferred for the subsequences. In our framework, the ball trajectory is represented as an image blob broken with cuts caused by occlusion or interference from a player. For those frame corresponding a cut, the ball is considered to become inv isible. To address the ambiguity in determining whether a cut is caused by temporary occlusion or player’s control, it is checked if the re-visible ball comes out in the tolerable range of the area where it is predicted by its dynamic model. If so, the cut is marked as temporary occlusion and virtually filled. For the resulted sub-blobs and cuts, events are extracted by maximum a posteriori estimation of the event makers. Experiments on sequences from two cameras shows satisfying results.