Spatio-temporal and object visualization in rugby union

Abstract

The use of computer software as an aid to rugby, and sports coaching in general, is becoming increasingly utilized. Videoed sport is the most widely used form of raw data for sports analysis, though it is currently not being used to its full potential. Patterns of player movement and position, both for individuals and groupings of players, are important for understanding the complexities of professional team sports, and yet are not being adequately addressed.

This paper outlines a project that aims to support coaching and/or commentary by visualizing and measuring the similarity of video-derived spatiotemporal information, and enabling timely access to relevant video clips. Specifically, methods by which a user of spatially-enabled sports software can visualize spatio-temporal and rugby object information will be discussed. Two issues are examined: (1) powerful spatio-temporal representation techniques for rugby constructs (such as the pitch, players and amalgamations of players: team, scrum, lineout, backline) and (2) user interface design and how it enables rugby object representation alongside the spatio-temporal visualization facility.