The Community for Technology Leaders
Green Image
Issue No. 03 - March (2009 vol. 21)
ISSN: 1041-4347
pp: 321-334
Heng Tao Shen , The University of Queensland, Brisbane
Jie Shao , The University of Queensland, Brisbane
Zi Huang , The University of Queensland, Brisbane
Xiaofang Zhou , The University of Queensland, Brisbane
Content-based video retrieval has been well investigated. However, despite the importance, few studies on video subsequence identification, which is to find the similar content to a short query clip from a long video sequence, have been published. This paper presents a graph transformation and matching approach to this problem, with extension to identify the occurrence of potentially different ordering, alignment or length due to content editing. With a batch query algorithm to retrieve similar frames, the mapping relationship between the query and the database video is first represented by a bipartite graph. The densely matched parts along the long sequence are then extracted, followed by a filter-and-refine search strategy to prune some irrelevant subsequences. During the filtering stage, Maximum Size Matching (MSM) is deployed for each subgraph constructed by the query and candidate subsequence to obtain a smaller set of candidates. During the refinement stage, Sub-Maximum Similarity Matching (SMSM) is devised to identify the subsequence, according to a robust video similarity model which incorporates visual content, temporal order, frame alignment and length information. The performance studies conducted on a long and diverse video recording validate our approach is promising in terms of both search accuracy and speed.
Multimedia databases, Search process, Information filtering

H. T. Shen, J. Shao, X. Zhou and Z. Huang, "Effective and Efficient Query Processing for Video Subsequence Identification," in IEEE Transactions on Knowledge & Data Engineering, vol. 21, no. , pp. 321-334, 2008.
96 ms
(Ver 3.3 (11022016))