基于流形正则协同训练模型的行为识别方法

doi:10.3976/j.issn.1002-4026.2018.01.019

Abstract

Abstract:

In this paper, a novel semi-supervised learning algorithm named Laplacian-regularized co-training(LapCo) was proposed. This method introduced Laplacian regularization to co-training model, and a large number of unlabeled sample data were used to train two classifiers from different view data, which could exchange unknown information between the two and update classifier to improve the recognition accuracy. In order to verify the effectiveness of the proposed algorithm, a large number of experiments were done on the action dataset UCF-iphone. The experimental results show that our proposed Laplacianregularized co-training model can effectively improve the accuracy of behavior recognition.

Key words: semi-supervised learning, co-training, Laplacian regularization, behavior recognition, manifold learning

CLC Number:

TP391

LIU Xiang-yang, LI Yang, JIANG Shu-ming，WANG Shuai. Manifold-regularized co-training model for behavior recognition[J].SHANDONG SCIENCE, 2018, 31(1): 116-.

References

［1］MOKHBER A, ACHARD C, MILGRAM M. Recognition of human behavior by spacetime silhouette characterization[J]. Pattern Recognition Letters, 2008, 29(1): 8189.
［2］POLAT E, YEASIN M, SHARMA R. Robust tracking of human body parts for collaborative human computer interaction[J]. Computer Vision and Image Understanding, 2003, 89(1): 4469.
［3］LEWIS C, BRAND C, CHERRY G, et al. Adapting user interface design methods to the design of educational activities[M]//Proceedings of the SIGCHI conference on human factors in computing systems. New York, NY, US:ACM Press/AddisonWesley Publishing Co., 1998: 619626.
［4］AYERS D, SHAH M. Monitoring human behavior from video taken in an office environment[J]. Image and Vision Computing, 2001, 19(12): 833846.
［5］周志华,王珏.机器学习及其应用[M].北京：清华大学出版社,2007.
［6］BLUM A, MITCHELL T. Combining labeled and unlabeled data with cotraining[M]//Proceedings of the eleventh annual conference on Computational learning theory. New York, NY, US：ACM, 1998: 92100.
［7］NIGAM K, GHANI R. Analyzing the effectiveness and applicability of cotraining[M]//Proceedings of the ninth international conference on Information and knowledge management. New York, NY, US：ACM, 2000: 8693.
［8］BALCAN M F, BLUM A, YANG K. Cotraining and expansion: Towards bridging theory and practice[C]// Proceedings of the 17th international conference on neural information processing systems. Cambridge, MA, US: MIT Press ,2004: 8996.
［9］SARKAR A. Applying cotraining methods to statistical parsing[C]//Proceedings of the second meeting of the North American chapter of the association for computational linguistics on language technologies. Stroudsburg, PA, US ：Association for Computational Linguistics, 2001: 18.
［10］PIERCE D, CARDIE C. Limitations of cotraining for natural language learning from large datasets[EB/OL].[20170304].. http://www.aclweb.org/anthology/W010501.
［11］ABNEY S.Bootstrapping[EB/OL].[20170304]. http://delivery.acm.org/10.1145/1080000/1073143/p360abney.pdf?ip=218.
57.145.1&id=1073143&acc=OPEN&key=4D4702B0C3E38B35%2E4D4702B0C3E38B35%2E4D4702B0C3E38B35%2E6D2181445
11F3437&CFID=841997444&CFTOKEN=14396075&__acm__=1513668499_91c8cfe6ab532cfec81740b5270b979c.
［12］LI Y, TAO D P, LIU W F, et al. Coregularization for classification[C]// Proceedings 2014 IEEE International Conference on Security, Pattern Analysis, and Cybernetics (SPAC). [S.l.]:IEEE, 2014: 218222.
［13］CORINNA C, VAPNIK V. Supportvector networks[J]. Machine learning, 1995, 20(3): 273297.
［14］BELKIN M, NIYOGI P. Laplacian eigenmaps and spectral techniques for embedding and clustering [EB/OL].[20170304]. http://papers.nips.cc/paper/1961laplacianeigenmapsandspectraltechniquesforembeddingandclustering.pdf .
［15］BELKIN M, NIYOGI P, SINDHWANI V. Manifold regularization: a geometric framework for learning from labeled and unlabeled examples[J]. Journal of Machine Learning Research, 2006,7:23992434.
［16］CHAPELLE O, SCH18LKOPF B, ZIEN A. Semisupervised learning using semidefinite programming[M]. Cambridge, MA, US: MIT Press, 2006.
［17］SHAHSHAHANI B M, LANDGREBE D A. The effect of unlabeled samples in reducing the small sample size problem and mitigating the Hughes phenomenon[J]. IEEE transactions on geoscience and remote sensing, 1994, 32(5): 10871095.
［18］WARDETZKY M，MATHUR S，KALBERER F，et al． Discrete laplace operators：No free lunch[EB/OL].[20170304]. http://www3.math.tuberlin.de/geometrie/Lehre/SS12/MathVisII/nofreelunch.pdf .
［19］WU Y, CHANG E Y, CHANG K C C, et al. Optimal multimodal fusion for multimedia data analysis[C]//Proceedings of the 12th annual ACM international conference on Multimedia. New York, NY, US： ACM, 2004: 572579.

Metrics

Comments

Recommended 0

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0), which permits third parties to freely share (i.e., copy and redistribute the material in any medium or format) and adapt (i.e., remix, transform, or build upon the material) the articles published in this journal, provided that appropriate credit is given, a link to the license is provided, and any changes made are indicated. The material may not be used for commercial purposes. For details of the CC BY-NC 4.0 license, please visit: https://creativecommons.org/licenses/by-nc/4.0

[1]	LIU Yu, GAO Shangbing, ZHANG Qintao, ZHANG Yingying. Object detection model YOLO-T for complex traffic scenarios [J]. Shandong Science, 2024, 37(6): 104-115.
[2]	LI Lian-wei,ZHANG Yuan-yu,YUE Zeng-you,XUE Cun-jin,FU Yu-xuan,XU Yang-feng. Extracting inland cage aquacultural areas from high-resolution remote sensing images using fully convolutional networks model [J]. Shandong Science, 2022, 35(2): 1-10.
[3]	XU Long,LI Yang. Smart contract-based access control architectural modeling [J]. Shandong Science, 2022, 35(1): 128-134.
[4]	MIAO Xiao, ZHANG Xin-dong, TANG Quan, YANG Si-yuan. Image restoration method based on partial differential equations [J]. Shandong Science, 2021, 34(2): 90-95.
[5]	WANG Yi-ming, JI Yong-gang. Analysis of the factors influencing target detection in sea clutter area through a compact high frequency surface wave radar [J]. Shandong Science, 2020, 33(6): 8-15.
[6]	YIN Su-ya, TANG Quan, ZHANG Xin. Improved image denoising method based on Perona-Malik model [J]. Shandong Science, 2020, 33(4): 124-130.
[7]	LIU Jie, LIU Yan, CAO Xuan, MA Ran, ZHANG Shu-wei, GUO Cui-lian. Weak signal detection technology of seawater nutrient salt [J]. Shandong Science, 2019, 32(6): 9-14.
[8]	CAO Lu-hui, DENG Yu-xiang, CHEN Tong, LI Zhao. A deep learning-based method for Chinese text-feature extraction and classification [J]. Shandong Science, 2019, 32(6): 106-111.
[9]	LI Xiang-chun, ZHANG Hao, LIU Xiao-yan, ZONG Fang-yi, LIU Jun-li. A method of underwater image enhancement based on transmittance optimization and color calibration [J]. Shandong Science, 2019, 32(2): 7-14.
[10]	LI Hui, XIE Wei-hao, LIU Shou-sheng, GAI Ying-ying. Realization of multiscale Retinex image enhancement algorithm based on GPU [J]. SHANDONG SCIENCE, 2017, 30(3): 103-109.
[11]	CHENG Yan,WU Bing-wei, ZHANG Ying-ying. Design of ocean interference detection system [J]. SHANDONG SCIENCE, 2016, 29(6): 19-23.
[12]	MA Ran, CAO Xuan,WANG Xiaohong, ZHANG Tianpeng, ZHANG Li. Binocular vision technology based monitoring technology of seawater biotoxicity [J]. SHANDONG SCIENCE, 2016, 29(5): 9-16.
[13]	SUN Jihua, ZHAO Yang, NAN Gangyang, MA Jian, LIU Shuai，JU Yang . Finite element based research on ultrasonic spectral characteristic of inner wall defect of metal materials [J]. SHANDONG SCIENCE, 2016, 29(5): 70-75.
[14]	JIA Tong-tong, ZHANG Xiao-le, SHI Yu-ying. Image blind deconvolution approach with modified regularization term [J]. SHANDONG SCIENCE, 2016, 29(3): 115-120.
[15]	FAN Zhongyong,ZHANG Zhijun,ZHANG Pengfei. Application of ontology technology in personalized recommendation system [J]. SHANDONG SCIENCE, 2016, 29(2): 101-105.

Manifold-regularized co-training model for behavior recognition

PDF (PC)

Like

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Metrics

Comments

Recommended 0