Improvement of pose recognition by sparse regularized convolutional neural network

Yuan Zhang; Xiao Yao; Zhongli Wang; Hao Su; Zihan Yu; Guanying Huo; Ning Xu; Xiaofeng Liu

doi:10.1109/RCAR47638.2019.9044112

Improvement of pose recognition by sparse regularized convolutional neural network

Yuan Zhang
, Xiao Yao^*
, Zhongli Wang
, Hao Su
, Zihan Yu
, Guanying Huo
, Ning Xu
, Xiaofeng Liu

^*Corresponding author for this work

College of Internet of Things Engineering
Beijing Jiaotong University
Hohai University Changzhou

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

The paper proposes a method of using deep model (CNN) with sparse regularization term to improve the performance of pose recognition. Convolutional neural network shows its limitations for pose recognition because of its bad convergence. It is believed that the activation function could be simplified by adding the sparseness term. Therefore, we present an algorithm applying the sparse regularization for the deep model with ReLU. Experimental results confirm that the proposed method can accelerate the convergence speed while a high recognition rate maintained.

Original language	English
Title of host publication	2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	248-251
Number of pages	4
ISBN (Electronic)	9781728137261
DOIs	https://doi.org/10.1109/RCAR47638.2019.9044112
State	Published - Aug 2019
Externally published	Yes
Event	2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019 - Irkutsk, Russian Federation Duration: 4 Aug 2019 → 9 Aug 2019

Publication series

Name	2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019

Conference

Conference	2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019
Country/Territory	Russian Federation
City	Irkutsk
Period	4/08/19 → 9/08/19

Access to Document

10.1109/RCAR47638.2019.9044112

Cite this

Zhang, Y., Yao, X., Wang, Z., Su, H., Yu, Z., Huo, G., Xu, N., & Liu, X. (2019). Improvement of pose recognition by sparse regularized convolutional neural network. In 2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019 (pp. 248-251). Article 9044112 (2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/RCAR47638.2019.9044112

Zhang, Yuan ; Yao, Xiao ; Wang, Zhongli et al. / Improvement of pose recognition by sparse regularized convolutional neural network. 2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 248-251 (2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019).

@inproceedings{5305dffd16954cba9e50c2b2e7fd48f2,

title = "Improvement of pose recognition by sparse regularized convolutional neural network",

abstract = "The paper proposes a method of using deep model (CNN) with sparse regularization term to improve the performance of pose recognition. Convolutional neural network shows its limitations for pose recognition because of its bad convergence. It is believed that the activation function could be simplified by adding the sparseness term. Therefore, we present an algorithm applying the sparse regularization for the deep model with ReLU. Experimental results confirm that the proposed method can accelerate the convergence speed while a high recognition rate maintained.",

author = "Yuan Zhang and Xiao Yao and Zhongli Wang and Hao Su and Zihan Yu and Guanying Huo and Ning Xu and Xiaofeng Liu",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019 ; Conference date: 04-08-2019 Through 09-08-2019",

year = "2019",

month = aug,

doi = "10.1109/RCAR47638.2019.9044112",

language = "英语",

series = "2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "248--251",

booktitle = "2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019",

address = "美国",

}

Zhang, Y, Yao, X, Wang, Z, Su, H, Yu, Z, Huo, G, Xu, N & Liu, X 2019, Improvement of pose recognition by sparse regularized convolutional neural network. in 2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019., 9044112, 2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019, Institute of Electrical and Electronics Engineers Inc., pp. 248-251, 2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019, Irkutsk, Russian Federation, 4/08/19. https://doi.org/10.1109/RCAR47638.2019.9044112

Improvement of pose recognition by sparse regularized convolutional neural network. / Zhang, Yuan; Yao, Xiao; Wang, Zhongli et al.
2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019. Institute of Electrical and Electronics Engineers Inc., 2019. p. 248-251 9044112 (2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Improvement of pose recognition by sparse regularized convolutional neural network

AU - Zhang, Yuan

AU - Yao, Xiao

AU - Wang, Zhongli

AU - Su, Hao

AU - Yu, Zihan

AU - Huo, Guanying

AU - Xu, Ning

AU - Liu, Xiaofeng

PY - 2019/8

Y1 - 2019/8

N2 - The paper proposes a method of using deep model (CNN) with sparse regularization term to improve the performance of pose recognition. Convolutional neural network shows its limitations for pose recognition because of its bad convergence. It is believed that the activation function could be simplified by adding the sparseness term. Therefore, we present an algorithm applying the sparse regularization for the deep model with ReLU. Experimental results confirm that the proposed method can accelerate the convergence speed while a high recognition rate maintained.

AB - The paper proposes a method of using deep model (CNN) with sparse regularization term to improve the performance of pose recognition. Convolutional neural network shows its limitations for pose recognition because of its bad convergence. It is believed that the activation function could be simplified by adding the sparseness term. Therefore, we present an algorithm applying the sparse regularization for the deep model with ReLU. Experimental results confirm that the proposed method can accelerate the convergence speed while a high recognition rate maintained.

UR - https://www.scopus.com/pages/publications/85083333462

U2 - 10.1109/RCAR47638.2019.9044112

DO - 10.1109/RCAR47638.2019.9044112

M3 - 会议稿件

AN - SCOPUS:85083333462

T3 - 2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019

SP - 248

EP - 251

BT - 2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019

Y2 - 4 August 2019 through 9 August 2019

ER -

Zhang Y, Yao X, Wang Z, Su H, Yu Z, Huo G et al. Improvement of pose recognition by sparse regularized convolutional neural network. In 2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 248-251. 9044112. (2019 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2019). doi: 10.1109/RCAR47638.2019.9044112

Improvement of pose recognition by sparse regularized convolutional neural network

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this