Multi-level Semantic Analysis for Sports Video

Semantic analysis is an active and interesting research topic in the field of sports video mining. In this chapter, the authors present a multi-level video semantic analysis framework that is featured by hybrid generative-discriminative probabilistic graphical models. A three-layer semantic space is proposed, by which the semantic video analysis is cast into two inter-related inference problems defined at different semantic levels. In the first stage, a multi-channel segmental hidden Markov model (MCSHMM) is developed to jointly detect multiple co-existent mid-level keywords from low-level visual features, which can serve as building blocks for high-level semantics. In the second stage, authors propose the auxiliary segmentation conditional random fields (ASCRFs) to discover the game flow from multi-channel key-words, which provides a unified semantic representation for both event and structure analysis. The use of hybrid generative-discriminative approaches in two different stages is proved to be effective and appropriate for multi-level semantic analysis in sports video. The experimental results from a set of American football video data demonstrate that the proposed framework offers superior results compared with other traditional machine learning-based video mining approaches.

Download Full-text

Semantic Analysis of Field Sports Video using a Petri-Net of Audio-Visual Concepts

The Computer Journal ◽

10.1093/comjnl/bxn058 ◽

2008 ◽

Vol 52 (7) ◽

pp. 808-823 ◽

Cited By ~ 3

Author(s):

L. Bai ◽

S. Lao ◽

A. F. Smeaton ◽

N. E. O'Connor ◽

D. Sadlier ◽

...

Keyword(s):

Petri Net ◽

Semantic Analysis ◽

Sports Video ◽

Visual Concepts ◽

Field Sports

Download Full-text

Automatic Multi-level Summarizations Generation Based on Basic Semantic Unit for Sports Video

Lecture Notes in Computer Science - PRICAI 2006: Trends in Artificial Intelligence ◽

10.1007/978-3-540-36668-3_56 ◽

2006 ◽

pp. 524-532

Author(s):

Chen Jianyun ◽

Zhao Xinyu ◽

Duan Miyi ◽

Wu Tingting ◽

Lao Songyang

Keyword(s):

Sports Video ◽

Multi Level ◽

Basic Semantic

Download Full-text

Semantic Analysis and Text Summarization in Socio-Technical Systems

Socio-Technical Decision Support in Air Navigation Systems - Advances in Mechatronics and Mechanical Engineering ◽

10.4018/978-1-5225-3108-1.ch008 ◽

2017 ◽

pp. 243-281 ◽

Cited By ~ 2

Author(s):

Nina Rizun

Keyword(s):

Decision Making ◽

Text Mining ◽

Semantic Analysis ◽

Text Summarization ◽

Technical System ◽

Synergy Effect ◽

Decision Making Processes ◽

Multi Level ◽

Technical Systems

In this chapter, the authors present the results of the development the text-mining methodology for increasing the reliability of the functioning of Socio-technical System (STS). Taking into account revealed strengths and weaknesses of Discriminant and Probabilistic approaches of Latent Semantic Relations analysis in of the abstracting and summarization projection, the Methodology of Two-level Single Document Summarization was developed. The Methodology assumes the following elements of novelty: based on obtaining a multi-level topical framework of the document (abstracting); uses the synergy effect of consistent usage the combination of two approaches for identification of conceptually significant elements of the text (summarization). The examples demonstrating the basic workability of proposed Methodology were presented. Such approaches should help human to increase the quality of supporting the decision-making processes of STS in real time.

Download Full-text

Event Detection in Sports Video Based on Generative-Discriminative Models

Computer Vision for Multimedia Applications ◽

10.4018/978-1-60960-024-2.ch009 ◽

2011 ◽

pp. 143-165

Author(s):

Guoliang Fan ◽

Yi Ding

Keyword(s):

Event Detection ◽

Semantic Analysis ◽

Building Blocks ◽

Semantic Space ◽

Sports Video ◽

Low Level ◽

Discriminative Models ◽

Video Mining ◽

High Level ◽

Different Levels

Semantic event detection is an active and interesting research topic in the field of video mining. The major challenge is the semantic gap between low-level features and high-level semantics. In this chapter, we will advance a new sports video mining framework where a hybrid generative-discriminative approach is used for event detection. Specifically, we propose a three-layer semantic space by which event detection is converted into two inter-related statistical inference procedures that involve semantic analysis at different levels. The first is to infer the mid-level semantic structures from the low-level visual features via generative models, which can serve as building blocks of high-level semantic analysis. The second is to detect high-level semantics from mid-level semantic structures using discriminative models, which are of direct interests to users. In this framework we can explicitly represent and detect semantics at different levels. The use of generative and discriminative approaches in two different stages is proved to be effective and appropriate for event detection in sports video. The experimental results from a set of American football video data demonstrate that the proposed framework offers promising results compared with traditional approaches.

Download Full-text