Walker-Independent Features for Gait Recognition from Motion Capture Data

  • Michal BalaziaEmail author
  • Petr Sojka
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10029)


MoCap-based human identification, as a pattern recognition discipline, can be optimized using a machine learning approach. Yet in some applications such as video surveillance new identities can appear on the fly and labeled data for all encountered people may not always be available. This work introduces the concept of learning walker-independent gait features directly from raw joint coordinates by a modification of the Fisher’s Linear Discriminant Analysis with Maximum Margin Criterion. Our new approach shows not only that these features can discriminate different people than who they are learned on, but also that the number of learning identities can be much smaller than the number of walkers encountered in the real operation.


Singular Value Decomposition Gait Cycle Radial Basis Function Neural Network Identity Class Correct Classification Rate 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.



Authors thank to the anonymous reviewers for their detailed commentary and suggestions. Data used in this project was created with funding from NSF EIA-0196217 and was obtained from Our extracted database is available at to support results reproducibility.


  1. 1.
    Ahmed, F., Paul, P.P., Gavrilova, M.L.: DTW-based kernel and rank-level fusion for 3D gait recognition using Kinect. Vis. Comput. 31(6), 915–924 (2015). CrossRefGoogle Scholar
  2. 2.
    Ahmed, M., Al-Jawad, N., Sabir, A.: Gait recognition based on Kinect sensor. Proc. SPIE 9139, 91390B–91390B-10 (2014). CrossRefGoogle Scholar
  3. 3.
    Ali, S., Wu, Z., Li, X., Saeed, N., Wang, D., Zhou, M.: Applying geometric function on sensors 3D gait data for human identification. In: Gavrilova, M.L., Kenneth Tan, C.J., Iglesias, A., Shinya, M., Galvez, A., Sourin, A. (eds.) Transactions on Computational Science XXVI. LNCS, vol. 9550, pp. 125–141. Springer, Heidelberg (2016). CrossRefGoogle Scholar
  4. 4.
    Andersson, V., Araujo, R.: Person identification using anthropometric and gait data from Kinect sensor (2015).
  5. 5.
    Balazia, M., Sojka, P.: Learning robust features for gait recognition by Maximum Margin Criterion. In: Proceedings of 23rd International Conference on Pattern Recognition, ICPR 2016, p. 6, December 2016. (preprint: arXiv:1609.04392)
  6. 6.
    Ball, A., Rye, D., Ramos, F., Velonaki, M.: Unsupervised clustering of people from ‘skeleton’ data. In: Proceedings of the Seventh Annual ACM/IEEE International Conference on Human-Robot Interaction, HRI 2012, pp. 225–226. ACM, New York (2012).
  7. 7.
    Castro, F.M., Marín-Jiménez, M.J., Guil, N., de la Blanca, N.P.: Automatic learning of gait signatures for people identification (2016). CoRR abs/1603.01006.
  8. 8.
    Choudhury, S.D., Tjahjadi, T.: Robust view-invariant multiscale gait recognition. Pattern Recogn. 48(3), 798–811 (2015). CrossRefGoogle Scholar
  9. 9.
    CMU Graphics Lab: Carnegie-Mellon Motion Capture (MoCap) Database (2003).
  10. 10.
    Dikovski, B., Madjarov, G., Gjorgjevikj, D.: Evaluation of different feature sets for gait recognition using skeletal data from Kinect. In: 37th International Convention on Information and Communication Technology, Electronics and Microelectronics, pp. 1304–1308, May 2014.
  11. 11.
    Kocsor, A., Kovács, K., Szepesvári, C.: Margin maximizing discriminant analysis. In: Boulicaut, J.-F., Esposito, F., Giannotti, F., Pedreschi, D. (eds.) ECML 2004. LNCS, vol. 3201, pp. 227–238. Springer, Heidelberg (2004). CrossRefGoogle Scholar
  12. 12.
    Kwolek, B., Krzeszowski, T., Michalczuk, A., Josinski, H.: 3D gait recognition using spatio-temporal motion descriptors. In: Nguyen, N.T., Attachoo, B., Trawiński, B., Somboonviwat, K. (eds.) ACIIDS 2014. LNCS, vol. 8398, pp. 595–604. Springer, Heidelberg (2014). CrossRefGoogle Scholar
  13. 13.
    Li, H., Jiang, T., Zhang, K.: Efficient and robust feature extraction by Maximum Margin Criterion. IEEE Trans. Neural Netw. 17(1), 157–165 (2006). CrossRefGoogle Scholar
  14. 14.
    Preis, J., Kessel, M., Werner, M., Linnhoff-Popien, C.: Gait recognition with Kinect. In: 1st International Workshop on Kinect in Pervasive Computing, New Castle, UK, pp. 1–4, 18–22 June 2012.
  15. 15.
    Reddy, V.R., Chakravarty, K., Aniruddha, S.: Person identification in natural static postures using Kinect. In: Agapito, L., Bronstein, M.M., Rother, C. (eds.) ECCV 2014 Workshops. LNCS, vol. 8926, pp. 93–808. Springer, Heidelberg (2015). Google Scholar
  16. 16.
    Sinha, A., Chakravarty, K., Bhowmick, B.: Person identification using skeleton information from Kinect. In: Proceedings of the Sixth International Conference on Advances in CHI, ACHI 2013, pp. 101–108 (2013).
  17. 17.
    Tafazzoli, F., Bebis, G., Louis, S.J., Hussain, M.: Genetic feature selection for gait recognition. J. Electron. Imaging 24(1), 013036 (2015). CrossRefGoogle Scholar
  18. 18.
    Vapnik, V.N.: The Nature of Statistical Learning Theory. Springer, New York (1995)CrossRefzbMATHGoogle Scholar
  19. 19.
    Zeng, W., Wang, C.: View-invariant gait recognition via deterministic learning. In: International Joint Conference on Neural Networks (IJCNN), pp. 3465–3472, July 2014.

Copyright information

© Springer International Publishing AG 2016

Authors and Affiliations

  1. 1.Faculty of InformaticsMasaryk UniversityBrnoCzech Republic

Personalised recommendations