一种基于多特征的视频人物聚类方法

检索一段视频中出现的人物并进行人物归类具有重要的研究意义和实用价值.本文提出一种基于多特征的视频人物检索聚类算法:先用一种结合人脸检测和物体跟踪的算法检测镜头人物,并提取人物衣服区域颜色以及声音作为人物特征,再用一种无监督模糊聚类方法对人物进行聚类,最后利用声音特征对聚类结果进行修正.该方法适用于人物数未知的条件下进行无监督的人物聚类.不同类型视频的试验证明该方法有效而实用.

Automated Person Indexing in Video

Person retrieval and indexing in video sequences are challenging task for many multimedia applications.This paper proposes a new method that indexes the person by their similarity.Firstly,the persons in a shot are detected and tracked through face detector and continuously adaptive mean shift algorithm.Then mid-level features such as clothes colors and voice are applied to represent the person.An unsupervised cluster method is performed to cluster the person for further index.At last,the cluster is validate...