A unified framework for semantic shot representation of sports video
Xiaofeng Tong, Qingshan Liu, Lingyu Duan, Hanqing Lu, Changsheng Xu, Qi Chuan Tian · 2005
The development of mid-level shot description helps to bridge the gap between low-level feature and high-level semantics in video indexing and analysis. In this paper, we present a unified framework for semantic shot representation in field-ball sports genres, in which a video shot is characterized via three essential properties, namely, camera shot size, subject in a scene and video production technology. The three properties clearly represent the primary factors of a shot, and provide a unified viewpoint of semantic shot definition. Based on this framework, we design an effective architecture for semantic shot management comprising three main components as: 1) flexible shot clustering and retrieval by adjusting the weights of three properties according to different requirements; 2) semantics based video temporal segmentation for further event recognition; and 3) comprehensive sports video semantics analysis. Extensive experiments on soccer, basketball and tennis demonstrate the effectiveness and validity of this framework.