An Optimized Algorithm on Multi-view Transform for Gait Recognition

  • Lingyun Chi
  • Cheng Dai
  • Jingren Yan
  • Xingang LiuEmail author
Conference paper
Part of the Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering book series (LNICST, volume 262)


Gait is one of the common used biometric features for human recognition, however, for some view angles, it is difficult to exact distinctive features, which leads to hindrance for gait recognition. Considering the challenge, this paper proposes an optimized multi-view gait recognition algorithm, which creates a Multi-view Transform Model (VTM) by adopting Singular Value Decomposition (SVD) on Gait Energy Image (GEI). To achieve the goal above, we first get the Gait Energy Image (GEI) from the gait silhouette data. After that, SVD is used to build the VTM, which can convert the gait view-angles to \( 90^\circ \) to get more distinctive features. Then, considering the image matrix is so large after SVD in practice, Principal Component Analysis (PCA) is used in our experiments, which helps to reduce redundancy. Finally, we measure the Euclidean distance between gallery GEI and transformed GEI for recognition. The experimental result shows that our proposal can significantly increase the richness of multi-view gait features, especially for angles offset to \( 90^\circ \).


Gait recognition Gait energy image View transform model Principal component analysis 



This work was supported by the Fundamental Research Funds for the Central Universities on the grant ZYGX2015Z009, and also supported by Applied Basic Research Key Programs of Science and Technology Department of Sichuan Province under the grant 2018JY0023.


  1. 1.
    Galar, M., et al.: A survey of fingerprint classification part I: taxonomies on feature extraction methods and learning models. Knowl. Based Syst. 81(C), 76–97 (2015)CrossRefGoogle Scholar
  2. 2.
    Jayakumari, V.V.: Face recognition techniques: a survey. World J. Comput. Appl. Technol. 1(2), 41–50 (2013)Google Scholar
  3. 3.
    De Marsico, M., Petrosino, A., Ricciardi, S.: Iris recognition through machine learning techniques: a survey. Pattern Recognit. Lett. 82(2), 106–115 (2016)Google Scholar
  4. 4.
    Liu, N., Lu, J., Tan, Y.-P.: Joint subspace learning for view-invariant gait recognition. IEEE Signal Process. Lett. 18(7), 431–434 (2011)CrossRefGoogle Scholar
  5. 5.
    Kusakunniran, W., et al.: A new view-invariant feature for cross-view gait recognition. IEEE Trans. Inf. Forensics Secur. 8(10), 1642–1653 (2013)CrossRefGoogle Scholar
  6. 6.
    Kwolek, B., Krzeszowski, T., Michalczuk, A., Josinski, H.: 3D gait recognition using spatio-temporal motion descriptor. In: Asian Conference on Intelligent Information and Database Systems, pp. 595–604. Springer, Cham (2014)CrossRefGoogle Scholar
  7. 7.
    Wolf, T., Babaee, M., Rigoll, G.: Multi-view gait recognition using 3d convolutional neural networks. In: IEEE International Conference on Image Processing, pp. 4165–4169. IEEE, USA (2016)Google Scholar
  8. 8.
    Makihara, Y., et al.: Gait recognition using a view transformation model in the frequency domain. In: European Conference on Computer Vision, pp. 151–163. Springer, Austria (2006)CrossRefGoogle Scholar
  9. 9.
    Kusakunniran, W., et al.: Multiple views gait recognition using view transformation model based on optimized gait energy image. In: IEEE International Conference on Computer Vision Workshops, pp. 1058–1064. IEEE, Japan (2010)Google Scholar
  10. 10.
    Han, J., Bhanu, B.: Individual recognition using gait energy image. IEEE Trans. Pattern Anal. Mach. Intell. 28(2), 316–322 (2005)CrossRefGoogle Scholar
  11. 11.
    Gu, J., Ding, X., Wang, S., et al.: Action and gait recognition from recovered 3-D human joints. IEEE Trans. Syst. Man Cybern. Part B 40(4), 1021–1033 (2010)CrossRefGoogle Scholar
  12. 12.
    Qing-Jiang, W.U.: Gait Recognition Based on PCA and SVM. Comput. Sci. (2006)Google Scholar
  13. 13.
    Yu, S., et al.: View invariant gait recognition using only one uniform model. In: International Conference on Pattern Recognition, pp. 889–894. IEEE, Mexico (2017)Google Scholar

Copyright information

© ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2019

Authors and Affiliations

  • Lingyun Chi
    • 1
  • Cheng Dai
    • 1
  • Jingren Yan
    • 1
  • Xingang Liu
    • 1
    Email author
  1. 1.School of Information and Communication EngineeringUniversity of Electronic Science and Technology of ChinaChengduChina

Personalised recommendations