Academic Journals Database
Disseminating quality controlled scientific knowledge

A new Method for Shot Identification in Basketball Video

ADD TO MY LIST
 
Author(s): Yun Liu | Xueying Liu | Chao Huang

Journal: Journal of Software
ISSN 1796-217X

Volume: 6;
Issue: 8;
Start page: 1468;
Date: 2011;
Original page

Keywords: mean-shift | shot identification | SURF | DBN

ABSTRACT
This paper presents semantic-based shot event identification. Based on Dynamic Bayesian Network (DBN), the gap between low-lever features and high-lever semantic will be resolve. We apply the mean-shift algorithm and Kalman filter to identify and track the ball and SURF(Speed up Robust Features) to find the basketball hoop. At last the DBN is applied to identify the shot events. Experimental results have shown our proposed method is effective for basketball event detection.
Affiliate Program      Why do you need a reservation system?