Please wait a minute...
FITEE
Front. Inform. Technol. Electron. Eng.  2010, Vol. 11 Issue (7): 525-537    DOI: 10.1631/jzus.C0910453
    
Multiple hypergraph ranking for video concept detection
Ya-hong Han, Jian Shao*, Fei Wu, Bao-gang Wei
Department of Computer Science and Technology, Zhejiang University, Hangzhou 310027, China
Multiple hypergraph ranking for video concept detection
Ya-hong Han, Jian Shao*, Fei Wu, Bao-gang Wei
Department of Computer Science and Technology, Zhejiang University, Hangzhou 310027, China
 全文: PDF 
摘要： This paper tackles the problem of video concept detection using the multi-modality fusion method. Motivated by
multi-view learning algorithms, multi-modality features of videos can be represented by multiple graphs. And the graph-based
semi-supervised learning methods can be extended to multiple graphs to predict the semantic labels for unlabeled video data.
However,  traditional  graphs  represent  only  homogeneous  pairwise  linking  relations,  and  therefore  the  high-order  correlations
inherent in videos, such as high-order visual similarities, are ignored. In this paper we represent heterogeneous features by multiple
hypergraphs and then the high-order correlated samples can be associated with hyperedges. Furthermore, the multi-hypergraph
ranking (MHR) algorithm is proposed by defining Markov random walk on each hypergraph and then forming the mixture Markov
chains  so  as  to  perform  transductive  learning  in  multiple  hypergraphs.  In  experiments  on  the  TRECVID  dataset,  a  triple-
hypergraph consisting of visual, textual features and multiple labeled tags is constructed to predict concept labels for unlabeled
video shots by the MHR framework. Experimental results show that our approach is effective.
关键词： Multiple hypergraph ranking;  Video concept detection;  Multi-view learning;  Multiple labeled tags;  Clustering    
Abstract: This paper tackles the problem of video concept detection using the multi-modality fusion method. Motivated by multi-view learning algorithms, multi-modality features of videos can be represented by multiple graphs. And the graph-based semi-supervised learning methods can be extended to multiple graphs to predict the semantic labels for unlabeled video data. However, traditional graphs represent only homogeneous pairwise linking relations, and therefore the high-order correlations inherent in videos, such as high-order visual similarities, are ignored. In this paper we represent heterogeneous features by multiple hypergraphs and then the high-order correlated samples can be associated with hyperedges. Furthermore, the multi-hypergraph ranking (MHR) algorithm is proposed by defining Markov random walk on each hypergraph and then forming the mixture Markov chains so as to perform transductive learning in multiple hypergraphs. In experiments on the TRECVID dataset, a triple-hypergraph consisting of visual, textual features and multiple labeled tags is constructed to predict concept labels for unlabeled video shots by the MHR framework. Experimental results show that our approach is effective.
Key words: Multiple hypergraph ranking    Video concept detection    Multi-view learning    Multiple labeled tags    Clustering
收稿日期: 2009-07-25 出版日期: 2010-07-06
CLC:  TP391  
基金资助: Project  supported  by  the  National  Natural  Science  Foundation  of China  (Nos.  60603096  and  60673088),  the  National  High-Tech  Re-
search  and  Development  Program  (863)  of  China  (No.  2006AA010107),  and  the  Program  for  Changjiang  Scholars  and  Innovative  Re-search Team in University of China (No. IRT0652) 
通讯作者: Jian SHAO     E-mail: jshao@zju.edu.cn
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章　　
Ya-hong Han
Jian Shao
Fei Wu
Bao-gang Wei

引用本文:

Ya-hong Han, Jian Shao, Fei Wu, Bao-gang Wei. Multiple hypergraph ranking for video concept detection. Front. Inform. Technol. Electron. Eng., 2010, 11(7): 525-537.

链接本文:

http://www.zjujournals.com/xueshu/fitee/CN/10.1631/jzus.C0910453    或    http://www.zjujournals.com/xueshu/fitee/CN/Y2010/V11/I7/525

[1] Rabia IRFAN , Sharifullah KHAN, Kashif RAJPOOT, Ali Mustafa QAMAR. TIE algorithm: a layer over clustering-based taxonomy generation for handling evolving data[J]. Front. Inform. Technol. Electron. Eng., 2018, 19(6): 763-782.
[2] Qin ZHANG, Guo-qiang ZHONG , Jun-yu DONG. An anchor-based spectral clustering method[J]. Front. Inform. Technol. Electron. Eng., 2018, 19(11): 1385-1396.
[3] A Ram CHOI, Sung Min KIM, Mee Young SUNG. Controlling the contact levels of details for fast and precise haptic collision detection[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(8): 1117-1130.
[4] Ke-shi GE, Hua-you SU, Dong-sheng LI, Xi-cheng LU. Efficient parallel implementation of a density peaks clustering algorithm on graphics processing unit[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(7): 915-927.
[5] Xin-zheng Xu, Shi-fei Ding, Zhong-zhi Shi, Hong Zhu. Optimizing radial basis function neural network based on rough sets and affinity propagation clustering algorithm[J]. Front. Inform. Technol. Electron. Eng., 2012, 13(2): 131-138.
[6] Jing Fan, Hai-feng Ji, Xin-xin Guan, Ying Tang. A GPU-based multi-resolution algorithm for simulation of seed dispersal[J]. Front. Inform. Technol. Electron. Eng., 2012, 13(11): 816-827.
[7] Suiang-Shyan Lee, Ja-Chen Lin. An accelerated K-means clustering algorithm using selection and erasure rules[J]. Front. Inform. Technol. Electron. Eng., 2012, 13(10): 761-768.
[8] Zhen-gong Cai, Xiao-hu Yang, Xin-yu Wang, Aleksander J. Kavs. A fuzzy formal concept analysis based approach for business component identification[J]. Front. Inform. Technol. Electron. Eng., 2011, 12(9): 707-720.
[9] Ji-ming Li, Yun-tao Qian. Clustering-based hyperspectral band selection using sparse nonnegative matrix factorization[J]. Front. Inform. Technol. Electron. Eng., 2011, 12(7): 542-549.
[10] Yan-xia Jin, Kai Zhang, James T. Kwok, Han-chang Zhou. Fast and accurate kernel density approximation using a divide-and-conquer approach[J]. Front. Inform. Technol. Electron. Eng., 2010, 11(9): 677-689.