Skip to main content

Advertisement

SpringerLink
Log in

Motion recognition for 3D human motion capture data using support vector machines with rejection determination

  • Published:
Multimedia Tools and Applications Aims and scope Submit manuscript

Abstract

This article presents a motion recognition strategy with rejection ability to extract the meaningful actions according to a given set of motion classes, or categories or types and reject such input patterns whose categories are not known. During the online recognition phrase, the multiple one-versus-one support vector machines are aggregated with the majority voting strategy over the most recent frames in a sliding window to predict the most probable type at each instance. And then, the corresponding index motion map is utilized to determine whether the predicted type should be accepted or not. The motion will be considered to be unknown when consecutive multiple frames are rejected. As a contribution, an adjusted self-organizing map algorithm is proposed to automatically learn the index motion map for each motion class, where the map size and topology are dynamically tuned by the intrinsic characteristics of the trained motions dataset. At the postprocessing step, the procedure is enhanced by an efficient key patterns-based verification strategy, which significantly improves the recognition precision. As a further contribution, we introduce a genetic algorithm learning algorithm to automatically learn the necessary key patterns for each class base on the previous learned index motion map. We evaluate our motion recognition model on various experiments conducted on synthetic data and real data from the freely available sets of motion capture database (HDM05). Experiment results show that the proposed strategy can not only classify motions correctly, but also identify the existence of unknown motion types.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15

Similar content being viewed by others

References

  1. Alahakoon D, Halgamuge SK, Srinivasan B (2000) Dynamic self-organizing maps with controlled growth for knowledge discovery. IEEE Trans Neural Netw 11:601–614

    Article  Google Scholar 

  2. Amit B, Philippe B, Chris D (2006) A support vector method for anomaly detection in hyperspectral imagery. IEEE Trans Geosci Remote Sens 44(8):2282–2291

    Article  Google Scholar 

  3. Arikan O, Forsyth DA, O’Brien J (2003) Motion synthesis from annotations. ACM Trans Graph 33(3):402–408

    Article  Google Scholar 

  4. Barbic J, Safonova A, Pan J-Y, et al. (2004) Segmenting motion capture data into distinct behaviors. Graph Interface, pp. 185–194

  5. Bertolami R, Zimmermann M, Bunke H (2006) Rejection strategies for offline handwritten text line recognition. Pattern Recogn Lett 27(16):2005–2012

    Article  Google Scholar 

  6. Blackmore J, Miikkulainen R (1993) Incremental grid growing: encoding high-dimensional structure into a two-dimensional feature map. Proc IEEE Int Conf Neural Netw (ICNN’93) 1:450–455

    Article  Google Scholar 

  7. Cao D, Masoud OT, Boley D, et al. (2004) Online motion classification using support vector machines. Proc IEEE Int Conf Robot Automat, pp. 2291–2296

  8. Chiu C, Chao S, Wu M et al (2004) Content-based retrieval for human motion data. J Vis Commun Image Represent 15(3):446–466

    Article  Google Scholar 

  9. Eiben AE, Smith JE (2007) Introduction to evolutionary computing. Springer

  10. Fritzke B (1994) Growing cell structures—a self-organizing network for unsupervised and supervised learning. Neural Netw 7:1441–1460

    Article  Google Scholar 

  11. Fritzke B (1995) Growing grid: a self-organizing network with constant neighborhood range and adaption strength. Neural Process Lett 2(5):1–5

    Article  Google Scholar 

  12. Furaoa S, Hasegawa O (2006) An incremental network for on-line unsupervised classification and topology learning. Neural Netw 19:90–106

    Article  Google Scholar 

  13. Gorski N (1997) Optimizing error-reject trade off in recognition systems. Proc ICDAR, pp. 1092–1096

  14. Guenterberg E, Yang AY, Ghasemzadeh H et al (2009) A method for extracting temporal parameters based on hidden markov models in body sensor networks with inertial sensors. IEEE Trans Inf Technol Biomed 13(6):1019–1030

    Article  Google Scholar 

  15. Guha T, Rabab K Ward. Learning sparse representations for human action recognition. IEEE Trans Pattern Anal Mach Intell, August, 34(8):1576–1588

  16. Guo K, Ishwar P, Konrad J (2007) Action recognition in video by sparse representation on covariance manifolds of silhouette tunnels. Tunnels ACCV, pp. 442–451

  17. Jiang M, Wang Z (2012) An incremental learning method based on probabilistic neural networks and adjustable fuzzy clustering for human activity recognition by using wearable sensors. IEEE Trans Inf Technol Biomed 16(4):691–699

    Article  Google Scholar 

  18. Kahol K, Tripathi P. Panchanathan S (2004) Automated gesture segmentation from dance sequences. Proc IEEE Int Conf Face Gesture Recogn, pp. 883–888

  19. Kahol K, Tripathi P, Panchanathan S, Rikakis T (2003) Gesture segmentation in complex motion sequences. Proc ICIP-03, pp. 105–108

  20. Koerich AL (2004) Rejection strategies for handwritten word recognition. Proc IWFHR, pp. 479–484

  21. Kohonen T (1988) Self-organized formation of topologically correct feature maps. Biol Cybern 43:59–69

    Article  Google Scholar 

  22. Krebel U (1999) Pairwise classification and support vector machines. In: Scholkops B, Burges CJC, Smola AJ (eds) Advances in Kernel methods: Support vector learning. MIT Press, Cambridge, pp 255–268

    Google Scholar 

  23. Landgrebe TCW, Tax DMJ, Paclík P, Duin RPW (2006) The interaction between classification and reject performance for distance-based reject-option classifiers. Pattern Recogn Lett 27(8):908–917

    Article  Google Scholar 

  24. Li C, Zheng SQ. Prabbakaran B (2007) Segmentation and recognition of motion streams by similarity search, ACM Transaction on Multimedia Computing. Commun Appl vol. 3, no.3, article No. 16

  25. Lv F, Nevatia R (2006) Recognition and segmentation of 3-D human action using HMM and multi-class AdaBoost. Proc ECCV, pp. 359–372

  26. Manabe S, Hatanaka T, Uosaki K, et al. (2006) Training hidden markov model structure with genetic algorithm for human motion pattern classification. Proc SICE-ICASE Int Joint Conf pp. 618–622

  27. Meinard Müller, Tido Röder (2006) Motion templates for automatic classification and retrieval of motion capture data, in Proceedings of the 2006 ACM Siggraph/Eurographics Symposium on Computer Animation, pp.137–146.

  28. Müller M, Baak A, Seidel H-P (2009) Efficient and robust annotation of motion capture data. Proc 2009 ACM Siggraph/Eurographics Symp Comput Animat pp. 17–26

  29. Muller M, Roer T, Clausen M, et al. (2007) Documentation: Mocap database HDM05, Computer graphics technical report CG-2007-2, University Bonn, June 2007. http://www.mpi-inf.mpg.de/resources/HDM05

  30. Pfau T, Ferrari M, Parsons K et al (2008) A hidden markov model based stride segmentation technique applied to equine inertial sensor trunk movement data. J Biomech 41(1):216–220

    Article  Google Scholar 

  31. Pitrelli J, Perrone MP (2002) Confidence modeling for verification post-processing for handwriting recognition. Proc IWFHR, pp. 30–35

  32. Pitrelli J, Perrone MP (2003) Confidence-scoring post-processing for off-line handwritten-character recognition verification. Proc ICDAR, pp. 278–282

  33. Qiu Q, Jiang Z, Chellappa R (2011) Sparse dictionary-based representation and recognition of action attributes. Comput Vis (ICCV), 2011 I.E. Int Conf, pp. 707–714

  34. Rizvi SA, Saasawi TN, Nasrabadi NM (2000) A clutter rejection technique for FLIR imagery using region based principal component analysis. Pattern Recogn 33(11):1931–1933

    Article  Google Scholar 

  35. Sakamoto Y, Kuriyama S, Kaneko T (2004) Motion map: image-based retrieval and segmentation of motion data. Proc 2004 ACM SIGGRAPH/Eurographics Symp Comput Animat, pp. 259–266

  36. Schölkopf B, Platt JC, Shawe-Taylor J, Smola AJ, Williamson RC (2001) Estimating the support of a high-dimensional distribution. Neural Comput 13(7):1443–1471

    Article  MATH  Google Scholar 

  37. Scholkopf B, Smola A (2002) Learning with kernels, support vector machines, regularization, optimization, and beyond. MIT Press

  38. Shen F, Hasegawa O (2007) An enhanced self-organizing incremental neural network for online unsupervised learning. Neural Netw 20:893–903

    Article  MATH  Google Scholar 

  39. van Kasteren T, Englebienne G, Kröse BJA (2010) An activity monitoring system for elderly care using generative and discriminative models. Pers Ubiquit Comput 14(6):489–498

    Article  Google Scholar 

  40. Vapnik V (1998) Statistical learning theory. Wiley, New York

    MATH  Google Scholar 

  41. Wang SB, Quattoni A, Morency L-P, Demirdjian D, Darrell T (2006) Hidden conditional random fields for gesture recognition. IEEE Comput Soc Conf Comput Vis Pattern Recogn (CVPR), pp. 1521–1527

  42. Wright J, Yang AY, Ganesh A et al (2009) Robust face recognition via sparse machine sparse representation. IEEE Trans Pattern Anal Mach Intell 31(2):210–227

    Article  Google Scholar 

  43. Wu S, Xia S, Wang Z et al (2009) Efficient motion data indexing and retrieval with local similarity measure of motion strings. Vis Comput 25(5–7):499–508

    Article  Google Scholar 

  44. Yang K, Sbababi C (2004) A PCA-based similarity measure for multivariate time series. Proc of the 2nd ACM Int Workshop Multimed Databases, pp. 65–74

  45. Zhao Q, Wang L, Horace HS Ip, et al. (2010) Human 3D motion recognition based on spatial-temporal context of joints. Int Conf Pattern Recogn (ICPR), pp. 2740–2743

  46. Zheng E-h, Zou C, Sun J, Chen L (2011) Cost-sensitive SVM with error cost and class-dependent reject cost. Int J Comput Theory Eng 3(1):130–135

    Article  Google Scholar 

Download references

Acknowledgments

The data used in this paper was obtained from HDM05. Thanks for their sharing. This research is partially supported by National Natural Science Funds of China (61173122), Key Project of Natural Science Foundation of Hunan Province, China (12JJ2038), Natural Science Foundation of Hunan Province, China (09JJ6102), The Postdoctoral Science Foundation of HuNan Academy of Science (2012RS4027), Social Science Foundation of HuNan province (12YBB286), The Scientific Research Fund of Hunan Provincial Education Department (12C0791).

Author information

Authors and Affiliations

  1. School of Information Science and Engineering, Central South University, Changsha, China

    Meiling Cai, Beiji Zou & Huanzhi Gao

  2. Electronic Information Engineering Department, Changsha Normal University, Changsha, China

    Meiling Cai

  3. Business School, Central South University, Changsha, China

    Juan Song

Authors
  1. Meiling Cai
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Beiji Zou
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Huanzhi Gao
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Juan Song
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Beiji Zou.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Cai, M., Zou, B., Gao, H. et al. Motion recognition for 3D human motion capture data using support vector machines with rejection determination. Multimed Tools Appl 70, 1333–1362 (2014). https://doi.org/10.1007/s11042-013-1749-5

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11042-013-1749-5

Keywords

Search

Navigation