No abstract

No abstract

Event detection is the key technology of content based sports video retrieval. Accurate detecting the boundaries of video events can effectively improve the user experience, and reduce the cost of video transmission. However, many existing works mainly focused on how to find the event, fewer on the determination of event boundary. In this work, we proposed an approach to determine the event boundary based on attackdefense transition analysis (ADTA) in soccer video. The main contributions of our work are as follows. A histogram based kickoff circle detection algorithm is proposed to facilitate the field zones partition. The proposed algorithm is fast and robust comparing with the existing methods. By the extracted mid-level features, the soccer playfield is divided into different zones, which imply different semantics in soccer game. Furthermore, the far view shot is segmented into sub-shots with captured zones, which benefit the highlight event and event boundary detection. The highlight events boundaries are accurately determined by attack-defense change points (ADCPs) detection combined with affection arousal curve. Experimental results demonstrated the effectiveness of the proposed approach.

As one of the key technologies in soccer video retrieval and summarization, highlight extraction has great academic and application value. According to the principle that the observer's affection state would fluctuate with the evolution of game process when watching soccer match video, a novel highlight extraction approach based on the improved affection arousal model is proposed. Compared with the existing works, our main contributions include the following. A novel feature -shot intensity is exploited to replace the motion activity, which greatly improves the computational performance of affection arousal model. Another helpful feature -replay factor is designed and successfully fused into the affection arousal model. This makes the affection arousal model reflect the variation of the true match process more accurately. In addition, event temporal transition pattern (ETTP) in soccer video is utilized to detect highlights boundaries effectively combined with the affection arousal curve. Experiments conducted on real-world soccer game videos have demonstrated the efficiency and effectiveness of the proposed approach.