作者: Tianlu Wang , Xianmin Zhang
DOI: 10.1007/978-3-642-37105-9_41
关键词:
摘要: This paper proposes an approach for trajectory matching in video retrieval. Algorithm is consist of three parts. First, a terse contains most important temporal and spatial characters are abstracted. Then, abstracted trajectories classified into several classes by using reformed k-means cluster method according to their position acceleration features. At the end, Gaussian Process regression model built trained clustered which similar with given retrieval targets found out. Advantages this algorithm include possibility generalized clustering different scales partial matching.