Until recently tactical analysis in elite soccer were based on observational data using variables which discard most contextual information. Analyses of team tactics require however detailed data from various sources including technical skill, individual physiological performance, and team formations among others to represent the complex processes underlying team tactical behavior. Accordingly, little is known about how these different factors influence team tactical behavior in elite soccer. In parts, this has also been due to the lack of available data. Increasingly however, detailed game logs obtained through next-generation tracking technologies in addition to physiological training data collected through novel miniature sensor technologies have become available for research. This leads however to the opposite problem where the shear amount of data becomes an obstacle in itself as methodological guidelines as well as theoretical modelling of tactical decision making in team sports is lacking. The present paper discusses how big data and modern machine learning technologies may help to address these issues and aid in developing a theoretical model for tactical decision making in team sports. As experience from medical applications show, significant organizational obstacles regarding data governance and access to technologies must be overcome first. The present work discusses these issues with respect to tactical analyses in elite soccer and propose a technological stack which aims to introduce big data technologies into elite soccer research. The proposed approach could also serve as a guideline for other sports science domains as increasing data size is becoming a wide-spread phenomenon.

/pdf/big-data-and-tactical-analysis-in-elite-soccer-future-2pu4o8bjc4.pdf

Big data and tactical analysis in elite soccer: future challenges and opportunities for sports science

Tracking and identifying players in sports videos filmed with a single pan-tilt-zoom camera has many applications, but it is also a challenging problem. This paper introduces a system that tackles this difficult task. The system possesses the ability to detect and track multiple players, estimates the homography between video frames and the court, and identifies the players. The identification system combines three weak visual cues, and exploits both temporal and mutual exclusion constraints in a Conditional Random Field (CRF). In addition, we propose a novel Linear Programming (LP) Relaxation algorithm for predicting the best player identification in a video clip. In order to reduce the number of labeled training data required to learn the identification system, we make use of weakly supervised learning with the assistance of play-by-play texts. Experiments show promising results in tracking, homography estimation, and identification. Moreover, weakly supervised learning with play-by-play texts greatly reduces the number of labeled training examples required. The identification system can achieve similar accuracies by using merely 200 labels in weakly supervised learning, while a strongly supervised approach needs a least 20,000 labels.

Learning to Track and Identify Players from Broadcast Sports Videos

Team-based invasion sports such as football, basketball and hockey are similar in the sense that the players are able to move freely around the playing area; and that player and team performance cannot be fully analysed without considering the movements and interactions of all players as a group. State of the art object tracking systems now produce spatio-temporal traces of player trajectories with high definition and high frequency, and this, in turn, has facilitated a variety of research efforts, across many disciplines, to extract insight from the trajectories. We survey recent research efforts that use spatio-temporal data from team sports as input, and involve non-trivial computation. This article categorises the research efforts in a coherent framework and identifies a number of open research questions.

Spatio-Temporal Analysis of Team Sports -- A Survey

Team-based invasion sports such as football, basketball, and hockey are similar in the sense that the players are able to move freely around the playing area and that player and team performance cannot be fully analysed without considering the movements and interactions of all players as a group. State-of-the-art object tracking systems now produce spatio-temporal traces of player trajectories with high definition and high frequency, and this, in turn, has facilitated a variety of research efforts, across many disciplines, to extract insight from the trajectories. We survey recent research efforts that use spatio-temporal data from team sports as input and involve non-trivial computation. This article categorises the research efforts in a coherent framework and identifies a number of open research questions.

Spatio-Temporal Analysis of Team Sports

A review of vision-based systems for soccer video analysis

Computer systems support many coaching activities performed before and after competitions, such as strategy development and performance evaluation, but competent assistance during games is much more demanding. It requires real-time interpretation of sensor data, the recognition and classification of ball actions, and fast-action game analysis and assessment. Only recently has high-precision microwave technology enabled computer systems to simultaneously track the position of players and the ball. Consequently, real-time game analysis systems must be able to automatically recognize intentional activities in a multiagent system with continually acting agents. To meet these demands, we've developed the football interaction and process model and a software system that can acquire, interpret, and analyze this model. Our FIPM system can acquire models of player skills, infer action-selection criteria, and determine player and team strengths and weaknesses.

https://ias.in.tum.de/_media/spezial/bib/bee05fipm.pdf

Computerized real-time analysis of football games

We propose automated sport game models as a novel technical means for the analysis of team sport games. The basic idea is that automated sport game models are based on a conceptualization of key notions in such games and probabilistically derived from a set of previous games. In contrast to existing approaches, automated sport game models provide an analysis that is sensitive to their context and go beyond simple statistical aggregations allowing objective, transparent and meaningful concept definitions. Based on automatically gathered spatio-temporal data by a computer vision system, a model hierarchy is built bottom up, where context-sensitive concepts are instantiated by the application of machine learning techniques. We describe the current state of implementation of the ASPOGAMO system including its computer vision subsystem that realizes the idea of automated sport game models. Their usage is exemplified with an analysis of the final of the soccer World Cup 2006.

/pdf/aspogamo-automated-sports-game-analysis-models-4gyeeal63g.pdf

ASpoGAMo: Automated Sports Game Analysis Models

This paper describes a camera-based observation system for football games that is used for the automatic analysis of football games and reasoning about multi-agent activity. The observation system runs on video streams produced by cameras set up for TV broadcasting. The observation system achieves reliability and accuracy through various mechanisms for adaptation, probabilistic estimation, and exploiting domain constraints. It represents motions compactly and segments them into classified ball actions.

/pdf/camera-based-observation-of-football-games-for-analyzing-kt64ix2lw1.pdf

Camera-based observation of football games for analyzing multi-agent activities

This paper describes ASPOGAMO, a visual tracking system that determines the coordinates and trajectories of football players in camera view based on TV broadcasts. To do so, ASPOGAMO solves a complex probabilistic estimation problem that consists of three subproblems that interact in subtle ways: the estimation of the camera direction and zoom factor, the tracking and smoothing of player routes, and the disambiguation of tracked players after occlusions. The paper concentrates on system aspects that make it suitable for operating under unconstrained conditions and in (almost) realtime. We report on results obtained in a public demonstration at RoboCup 2006 where we conducted extensive experiments with real data from live coverage of World Cup 2006 games in Germany.

/pdf/visually-tracking-football-games-based-on-tv-broadcasts-4udsh7it4g.pdf

Visually tracking football games based on TV broadcasts

In this paper we present ASPOGAMO, a vision system capable of es- timating motion trajectories of soccer players taped on video The system per- forms well in a multitude of application scenarios because of its adaptivity to various camera setups, such as single or multiple camera settings, static or dy- namic ones Furthermore, ASPOGAMO can directly process image streams taken from TV broadcast, and extract all valuable information despite scene interrup- tions and cuts between different cameras The system achieves a high level of robustness through the use of modelbased vision algorithms for camera estima- tion and player recognition and a probabilistic multi-player tracking framework capable of dealing with occlusion situations typical in team-sports The continu- ous interplay between these submodules is adding to both the reliability and the efficiency of the overall system

/pdf/an-adaptive-vision-system-for-tracking-soccer-players-from-188x05et9w.pdf

Bernhard Kirchlechner

Papers

Computerized real-time analysis of football games

ASpoGAMo: Automated Sports Game Analysis Models

Camera-based observation of football games for analyzing multi-agent activities

Visually tracking football games based on TV broadcasts

An Adaptive Vision System for Tracking Soccer Players from Variable Camera Settings