Combining Content and Context Information Fusion for Video Classification and Retrieval
Content-Based Video Retrieval bas been a challenging problem und its performance relies on the modeling and representation of the video data and the underlying similarity metric. Most existing metrics evaluate pairwise shot similarity based only on shot perceptual content, whicb is denoted as content-based similarity. In this study, our concern is to recognize and detect video events that are semantically similar. Thus, we extend the contentbased similarity to measure the conceptual content of shots. Here, conceptual content refers to the dynamic semantic concept which reflects a human-action regardless the perceptual/visual appearance In addition, we propose a new similarity metric to make use of the shot contexts in video clips collection. The context of a shot is built by constructing a vector with each dimension representing the content similarity between the shot and any shot in the video collection. The context similari0 between two videos is obtained by computing the similarity between the corresponding context vectors using the vector similarity functions. Furthermore, a linear and non linear fusion scbemes are introduced to compute the relative contributions of each similarity in the overall retrieval and classification process. Experimental results demonstrate that the use of the context similarity can significantly improve the retrieval performance.
Retrieval Similarity Measures Context Content
Bashar Tahayna Saadat Alhashmi Yandan Wang Khaled Abbas
School of ITMonash University Malaysia Faculty of Computer Science and IT University of Malaya Malaysia
国际会议
2010 2nd International Conference on Signal Processing System(2010年信号处理系统国际会议 ICSPS 2010)
大连
英文
1440-1444
2010-07-05（万方平台首次上网日期，不代表论文的发表时间）