Attention modeling of game excitement for sports videos

Abstract

This paper presents a novel attention modeling method that combines the visual attention features with the contextual game status information for sports videos. Two critical issues for attention-based video content analysis are addressed in this paper. (1) It illustrates the approach of extracting the visual attention map and illustrates the algorithm for determining the contextual excitement value. Semantic contextual inference is used to simulate how the video content attracts the subscribers. (2) It presents the fusion methodology of visual and contextual attention analysis based on the characteristics of human excitement. The experimental results demonstrate the efficiency and the robustness of our system by means of some baseball game videos. © 2011 Springer-Verlag.