IEEE TRANSACTIONS ON MULTIMEDIA, VOL. 11, NO. 1, JANUARY 2009 49
Event Tactic Analysis Based on
Broadcast Sports Video
Guangyu Zhu, Changsheng Xu, Senior Member, IEEE, Qingming Huang, Member, IEEE,
Yong Rui, Senior Member, IEEE, Shuqiang Jiang, Member, IEEE, Wen Gao, Fellow, IEEE, and
Hongxun Yao, Member, IEEE
Abstract—Most existing approaches on sports video analysis
have concentrated on semantic event detection. Sports profes-
sionals, however, are more interested in tactic analysis to help
improve their performance. In this paper, we propose a novel
approach to extract tactic information from the attack events in
broadcast soccer video and present the events in a tactic mode
to the coaches and sports professionals. We extract the attack
events with far-view shots using the analysis and alignment of
web-casting text and broadcast video. For a detected event, two
tactic representations, aggregate trajectory and play region se-
quence, are constructed based on multi-object trajectories and
field locations in the event shots. Based on the multi-object trajec-
tories tracked in the shot, a weighted graph is constructed via the
analysis of temporal–spatial interaction among the players and
the ball. Using the Viterbi algorithm, the aggregate trajectory is
computed based on the weighted graph. The play region sequence
is obtained using the identification of the active field locations in
the event based on line detection and competition network. The
interactive relationship of aggregate trajectory with the infor-
mation of play region and the hypothesis testing for trajectory
temporal–spatial distribution are employed to discover the tactic
patterns in a hierarchical coarse-to-fine framework. Extensive
experiments on FIFA World Cup 2006 show that the proposed
approach is highly effective.
Index Terms—Event detection, object tracking, trajectory anal-
ysis, tactic analysis, sports video analysis.
Manuscript received September 05, 2007; revised February 04, 2008.
Current version published January 08, 2009. This work was supported in part
by National Natural Science Foundation of China under Grants 60773136 and
60702035, in part by National Hi-Tech Development Program (863 Program)
of China under Grant 2006AA01Z117, and in part by “Science100 Program” of
Chinese Academy of Sciences under Grant 99T3002T03. The associate editor
coordinating the review of this manuscript and approving it for publication was
Dr. Alan Hanjalic.
G. Zhu and H. Yao are with the School of Computer Science and Tech-
nology, Harbin Institute of Technology, Harbin, China, 150001 (e-mail:
gyzhu@jdl.ac.cn; yhx@vilab.hit.edu.cn).
C. Xu is with the National Lab of Pattern Recognition, Institute of Au-
tomation, Chinese Academy of Sciences, Beijing 100080, China and also
with the China-Singapore Institute of Digital Media, Singapore (e-mail:
csxu@nlpr.ia.ac.cn).
Q. Huang is with the Graduate School of Chinese Academy of Sciences, Bei-
jing, China, 100039 (e-mail: qmhuang@jdl.ac.cn).
Y. Rui is with the Microsoft China R&D (CRD) Group, Beijing, China,
100080 (e-mail: yongrui@microsoft.com).
S. Jiang is with the Key Lab of Intelligent Information Processing, Chinese
Academy of Sciences, Beijing, China, 100190 (e-mail: sqjiang@jdl.ac.cn).
W. Gao is with the School of Computer Science and Technology, Harbin Insti-
tute of Technology, Harbin, China, 150001 and also with the Institute of Digital
media, Peking University, Beijing 100871, China (e-mail: wgao@jdl.ac.cn).
Color versions of one or more figures in this paper are available online at
http://ieeexplore.ieee.org).
Digital Object Identifier 10.1109/TMM.2008.2008918
I. INTRODUCTION
S
PORTS content is expected to be a key driver for com-
pelling new infotainment applications and services
because of its mass appeal and inherent structures which are
amenable for automatic processing. Due to its wide viewership
and tremendous commercial value, there has been an explosive
growth in the research area of sports video analysis [1]–[21].
From a sports-watcher point of view, only some portions in
a sports video are worth viewing. These video segments of
interest are the semantic events which have certain high-level
concepts, such as goals in soccer games and homeruns in
baseball games. The detection and extraction of game events
can be achieved by semantic analysis of sports video, to which
most of current research efforts have been devoted [1]–[11].
Semantic analysis aims at detecting and extracting informa-
tion that describes “facts” in a video, e.g., the “goal” events of
a soccer match. In contrast, tactic analysis of sports video aims
to recognize and discover tactic patterns and match strategies
that teams or individual players used in the games. From the
coach and sports professional point of view, they are more inter-
ested in the tactic strategies in the specific game events. Taking
soccer game as an example, there is a great interest from the
coaches and players in better understanding the process and pat-
terns of attacks so that he/she is able to improve the team perfor-
mance during the game and better adapt the training plan. Fur-
thermore, soccer fans, especially the hardcore ones, may also
be interested in the results from tactic perspective for enjoying
soccer games with the additional information beyond traditional
event “facts.” Unfortunately, existing semantic approaches on
sports video usually only summarize the extracted events and
then present to the users directly without any further analysis on
the tactics. Today, for sports professionals to obtain the results
of tactic analysis, it is common for them to employ people to
conduct the analysis manually. This process is labor-intensive,
time-consuming and error-prone. Consequently, there exists a
compelling case to automate sports tactic analysis. However, to
the best of knowledge, immediately related work in the field is
very limited.
In this paper, we propose a novel tactic analysis approach on
broadcast soccer games. As the most representative genre of the
team sports, the soccer game tends to follow the trend of a group
cooperation of players from the tactic perspective to compete
with the opponent team and to achieve the final goals. Tactic
analysis and summarization for the soccer game can potentially
offer assistance to the coaches, players and professionals on im-
proving their own skills and studying the opponent strategies.
1520-9210/$25.00 © 2009 IEEE
Authorized licensed use limited to: IEEE Xplore. Downloaded on March 9, 2009 at 05:53 from IEEE Xplore. Restrictions apply.