基于“bag of words”的视频匹配方法

被引:13
作者
李远宁
刘汀
蒋树强
黄庆明
机构
[1] 中国科学院计算技术研究所
关键词
视频匹配; 视觉关键词; bagofwords; 倒排索引;
D O I
暂无
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
提出了一种利用"bag of words"模型对视频内容进行建模和匹配的方法。通过量化视频帧的局部特征构建视觉关键词(visual words)辞典,将视频的子镜头表示成若干视觉关键词的集合。在此基础上构建基于子镜头的视觉关键词词组的倒排索引,用于视频片段的匹配和检索。这种方法保留了局部特征的显著性及其相对位置关系,而且有效地压缩了视频的表达,加速的视频的匹配和检索过程。实验结果表明,和已有方法相比,基于"bag of words"的视频匹配方法在大视频样本库上获得了更高的检索精度和检索速度。
引用
收藏
页码:147 / 151
页数:5
相关论文
共 7 条
[1]  
A bayesian hierarchical model for learning natural scene categories. LI F F,PERONA P. IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR) . 2005
[2]  
Object recognition from local scale invariant feature. DAVID L. Inter-national Conference on Computer Vision(IEEE ICCV) . 1999
[3]  
Using dependent regions for object categorization in a generative framework. WANG G,ZHANG Y,LI F F. IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR) . 2006
[4]  
Discovering objects and their location in images. SIVIC J,RUSSELL B,EFROS A A. International Conference on Computer Vision(IEEE ICCV) . 2005
[5]  
A fast video clip retrieval algorithm based on VA-File. LIU F,DONG D,XUE X Y. SPIE Storage and Retrieval for Media Database . 2004
[6]  
Efficient video retrieval using shot clustering and alignment. Nai-xiang Lian,Yap-Peng Tan,Kap Luk Chan. The 4th Int’l Conf. Information, Communications & Signal Processing and Fourth Pacific-Rim Conf on Multimedia . 2003
[7]  
Efficient video similarity measurement with video signature. S S Cheung,A Zakhor. IEEE Transactions on Circuits and Systems for Video Technology . 2003