Database Reference
In-Depth Information
The Highlights-On-Demand for a game needs a system that can analyze the contents
of the broadcast and derive the semantics from the input videos. These semantics
can be made available to the users for querying in order to create a true On-Demand
experience. This section addresses this issue by presenting an application system to
classify American Football (NFL) video shots, using MPEG-7 motion descriptors,
and enhancing the indexing capabilities of the system with MPEG-7 audio and Mel
Frequency Cepstrum Coefficients (MFCC) features.
Recently, some research has been conducted on automating the process of
indexing and annotating sports video streams. Nearly all the major sports have been
used to test the indexing and retrieval systems. One of the major projects working
on the generation of semantic sports video annotations is the ASSAVID project. As
detailed in [
213
], this project focuses on developing a system that can categorize
different types of sports and provides users with an interface to query events in a
particular sport.
In [
214
], audio, textual and visual information are used to classify NFL video
into events like touchdowns and field goals. In [
215
], different types of formations
within NFL games were classified using the natural language commentary from
the game, the geometrical information about the play and the domain knowledge.
In [
216
], closed caption text and audio visual information were utilized to classify
plays into three categories namely: scrimmage, FG/XP and K/P.
The aforementioned works rely on domain knowledge to classify different high
level concepts within American football. On the other hand, the video classification
system in the current work classifies recurring events of the game without using any
domain knowledge. These recurring events are the most basic components of the
game. By classifying these basic components first we can look for higher concepts
contained within each of the basic events and thus generate a hierarchical graph
of concepts which varies from low level to high level. The standard descriptors of
MEPG-7 are utilized as the basic feature set. In [
217
], the author shares proposed
applications for generating summary highlights in the sports domain using MPEG-7
motion descriptors, but MPEG-7 audio and motion descriptors have not been used
to index recurring events in the American football domain.
7.6.1
Localization of Play Events
Sports have a very well defined structure. They have a set of rules that must be
followed in order for the game to be played properly. Many sports such as golf,
baseball, bowling and American football have a requirement that the team or players
must be in a distinctive position before each play begins. In golf, the player positions
himself by the ball in order to hit it in a certain direction. Likewise in American
football, the two teams first line up face to face before the ball is snapped to begin
the play. The common theme among all these sports is that before the play starts,
the level of motion activity in the video is lower compared to when the play has
started. This distinction in the motion activity is utilized in the proposed algorithm
to segment play events from non-play events. Figure
7.21
shows the magnitude of
motion vectors in different types of NFL plays.
Search WWH ::
Custom Search