Machine Learning-Based Soccer Video Summarization System.

Citation:
Hassanien, A. E., "Machine Learning-Based Soccer Video Summarization System.", Multimedia, Computer Graphics and Broadcasting - International Conference, MulGraB 2011,, Jeju Island, Korea, December 8-10, 2011.

Date Presented:

December 8-10

Abstract:

This paper presents a machine learning (ML) based event detection and summarization system for soccer matches. The proposed system is composed of six phases. Firstly, in the pre-processing phase, the system segments the whole video stream into small video shots. Then, in the shot processing phase, it applies two types of classification to the video shots resulted from the pre-processing phase. Afterwards, in the replay detection phase, the system applies two machine learning algorithms, namely; support vector machine (SVM) and neural network (NN), for emphasizing important segments with logo appearance. Also, in the score board detection phase, the system uses both ML algorithms for detecting the caption region providing information about the score of the game. Subsequently, in the excitement event detection phase, the system uses k-means algorithm and Hough line transform for detecting vertical goal posts and Gabor filter for detecting goal net. Finally, in the logo-based event detection and summarization phase, the system highlights the most important events during the match. Experiments on real soccer videos demonstrate encouraging results. Compared to the performance results obtained using SVM classifier, the proposed system attained good NN-based performance results concerning recall ratio, however it attained poor NN-based performance results concerning precision ratio.

Related External Link

Tourism