Full Text:   <2965>

CLC number: TP37

On-line Access: 

Received: 2006-05-10

Revision Accepted: 2006-10-15

Crosschecked: 0000-00-00

Cited: 0

Clicked: 5788

Citations:  Bibtex RefMan EndNote GB/T7714

-   Go to

Article info.
Open peer comments

Journal of Zhejiang University SCIENCE A 2006 Vol.7 No.12 P.2063-2072


Ensemble learning HMM for motion recognition and retrieval by Isomap dimension reduction

Author(s):  XIANG Jian, WENG Jian-guang, ZHUANG Yue-ting, WU Fei

Affiliation(s):  School of Computer Science, Zhejiang University, Hangzhou 310027, China

Corresponding email(s):   xiang_bj@cs.zju.edu.cn, wengjg@cs.zju.edu.cn

Key Words:  Feature, Isomap, HMM (hidden Markov model), Ensemble learning, Motion recognition and retrieval

XIANG Jian, WENG Jian-guang, ZHUANG Yue-ting, WU Fei. Ensemble learning HMM for motion recognition and retrieval by Isomap dimension reduction[J]. Journal of Zhejiang University Science A, 2006, 7(12): 2063-2072.

@article{title="Ensemble learning HMM for motion recognition and retrieval by Isomap dimension reduction",
author="XIANG Jian, WENG Jian-guang, ZHUANG Yue-ting, WU Fei",
journal="Journal of Zhejiang University Science A",
publisher="Zhejiang University Press & Springer",

%0 Journal Article
%T Ensemble learning HMM for motion recognition and retrieval by Isomap dimension reduction
%A WENG Jian-guang
%A ZHUANG Yue-ting
%A WU Fei
%J Journal of Zhejiang University SCIENCE A
%V 7
%N 12
%P 2063-2072
%@ 1673-565X
%D 2006
%I Zhejiang University Press & Springer
%DOI 10.1631/jzus.2006.A2063

T1 - Ensemble learning HMM for motion recognition and retrieval by Isomap dimension reduction
A1 - XIANG Jian
A1 - WENG Jian-guang
A1 - ZHUANG Yue-ting
A1 - WU Fei
J0 - Journal of Zhejiang University Science A
VL - 7
IS - 12
SP - 2063
EP - 2072
%@ 1673-565X
Y1 - 2006
PB - Zhejiang University Press & Springer
ER -
DOI - 10.1631/jzus.2006.A2063

Along with the development of motion capture technique, more and more 3D motion databases become available. In this paper, a novel approach is presented for motion recognition and retrieval based on ensemble HMM (hidden Markov model) learning. Due to the high dimensionality of motion’s features, isomap nonlinear dimension reduction is used for training data of ensemble HMM learning. For handling new motion data, isomap is generalized based on the estimation of underlying eigenfunctions. Then each action class is learned with one HMM. Since ensemble learning can effectively enhance supervised learning, ensembles of weak HMM learners are built. Experiment results showed that the approaches are effective for motion data recognition and retrieval.

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article


[1] Assa, J., Caspi, Y., Cohen-Or, D., 2005. Action synopsis: Pose selection and illustration. Proceedings of ACM SIGGRAPH, 24(3):667-676.

[2] Bengio, Y., Paiement, J.F., Vincent, P., 2003. Out-of-Sample Extensions for LLE, ISOMAP, MDS, Eigenmaps, and Spectral Clustering. Proceedings of Neural Information Processing Systems, 16:177-184.

[3] Beyer, K., Goldstein, J., Ramakrishnan, R., Shaft, U., 1999. When is “Nearest Neighbor” Meaningful? Proceedings of the International Conference on Database Theory, p.217-235.

[4] Breiman, L., 1996. Bagging predictors. Machine Learning, 24(2):123-140.

[5] Chiu, Y., Chao, S.P., Wu, M.Y., Yang, S.N., Lin, H.C., 2004. Content-based retrieval for human motion data. Journal of Visual Communication and Image Representation, 15(3):446-466.

[6] Freund, Y., Schapire, R.E., 1995. A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting. Proceedings of the 2th European Conference on Computational Learning Theory, p.23-37.

[7] He, X.F., Ma, W.Y., Zhang, H.J., 2004. Learning an Image Manifold for Retrieval. Proceedings of 12th ACM International Conference on Multimedia, p.17-23.

[8] Law, M.H.C., Zhang, N., Jain, A.K., 2004. Nonlinear Manifold Learning for Data Stream. Proceedings of SIAM Data Mining, p.33-44.

[9] Leung, E.W.C., Li, Q., 2002. Media-on-Demand for Agent-Based Collaborative Tutoring Systems on the Web. Proceedings of the IEEE Pacific Rim Conference on Multimedia, p.976-984.

[10] Liu, F., Zhuang, Y.T., Wu, F., Pan, Y.H., 2003. 3D motion retrieval with motion index tree. Computer Vision and Image Understanding, 92(2-3):265-284.

[11] Liu, X.M., Chen, T., 2004. Video-Based Face Recognition Using Adaptive Hidden Markov Models. Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition, p.340-345.

[12] Lv, F.J., Nevatia, R., 2006. Recognition and Segmentation of 3-D Human Action Using HMM and Multi-class AdaBoost. 9th European Conference on Computer Vision (ECCV), 3954:359-372.

[13] Mueller, M., Roeder, T., Clausen, M., 2005. Efficient content-based retrieval of motion capture data. Proceedings of ACM SIGGRAPH, 24(3):677-685.

[14] Rabiner, L., 1989. A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition. Proceedings of the IEEE, p.257-286.

[15] Roweis, S.T., Saul, L.K., 2000. Nonlinear dimensionality reduction by locally linear embedding. Science, 290(5500):2323-2326.

[16] Russell, S., Norvig, P., 2002. Artificial Intelligence: A Modern Approach (Second Ed.). Prentice Hall, Englewood Cliffs, New Jersey.

[17] Shi, L.K., He, P.L., Liu, B., 2005. A Robust Generalization of Isomap for New Data. Proceedings of the Fourth International Conference on Machine Learning and Cybernetics, p.1702-1712.

[18] Starner, T., 1995. Visual Recognition of American Sign Language Using Hidden Markov Models. Master’s Thesis. MIT Media Laboratory, p.189-194.

[19] Tenenbaum, J., Silva, V.D., Langford, J., 2000. A global geometric framework for nonlinear dimensionality reduction. Science, 290(5500):2319-2323.

[20] Yin, P., Essa, I., Rehg, J.M., 2004. Asymmetrically Boosted HMM for Speech Reading. Proceedings of the IEEE Computer Society Conference, p.755-761.

[21] Zhai, J., Yang, J., Li, Q., Liu, W.Y., Feng, B., 2003. Rich Media Retrieval on the Web—A Multi-level Indexing Approach. Proceedings of the 12th International World Wide Web Conference, p.383-390.

Open peer comments: Debate/Discuss/Question/Opinion


Please provide your name, email address and a comment

Journal of Zhejiang University-SCIENCE, 38 Zheda Road, Hangzhou 310027, China
Tel: +86-571-87952783; E-mail: cjzhang@zju.edu.cn
Copyright © 2000 - 2024 Journal of Zhejiang University-SCIENCE