A Cost Effective Method for Matching the 3D Motion Trajectories

Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 215)

Abstract

3D trajectory data have progressively become common since more devices which are possible to acquire motion data were produced. These technology advancements promote studies of motion analysis based on the 3D trajectory data. Even though similarity measurement of trajectories is one of the most important tasks in 3D motion analysis, existing methods are still limited. Recent researches focus on the full length 3D trajectory data set. However, it is not true that every point on the trajectory plays the same role and has the same meaning. In this situation, we developed a new cost effective method that uses the feature ‘signature’ which is a flexible descriptor computed only from the region of ‘elbow points’. Therefore, our proposed method runs faster than other methods which use the full length trajectory information. The similarity of trajectories is measured based on the signature using an alignment method such as dynamic time warping (DTW), continuous dynamic time warping (CDTW) or longest common subsequence (LCSS) method. In the experimental studies, we compared our method with two other methods using Australian sign word dataset to demonstrate the effectiveness of our algorithm.

Keywords

Motion analysis 3D trajectory Similarity of trajectory Sign words 

Notes

Acknowledgments

This study was financially supported by Chonnam National University, 2011

References

  1. 1.
    Croitoru A, Agouris P, Stefanidis A (2005) 3D trajectory matching by pose normalization. In: Proceedings of the 13th annual ACM international workshop on geographic information systems, pp 153–162Google Scholar
  2. 2.
    Wu S, Li YF (2009) Flexible signature descriptions for adaptive motion trajectory representation, perception and recognition. Pattern Recognit 42:194–214 Google Scholar
  3. 3.
    Yang JY, Li YF (2010) A new descriptor for 3D trajectory recognition. Automation and logistics (ICAL), pp 37–42Google Scholar
  4. 4.
    Vlachos M, Hadjieleftheriou M, Gunopulos D, Keogh E (2003) Indexing multi-dimensional time-series with support for multiple distance measures. In: Proceedings of the ninth ACM SIGKDD international conference on knowledge discovery and data mining, pp 216–225Google Scholar
  5. 5.
    Vlachos M, Kollios G, Gunopulos D (2008) Discovering similar multidimensional trajectories. In: Proceedings 18th international conference on data engineering, pp 673–684Google Scholar
  6. 6.
    Aach J, Church G (2001) Aligning gene expression time series with time warping algorithms. Bioinformatics 17:495–508CrossRefGoogle Scholar
  7. 7.
    Kehtarnavaz N, deFigueiredo JP (1988) A 3D contour segmentation scheme based on curvature and torsion. IEEE Trans Pattern Anal Mach Intell 10(5):707–713Google Scholar
  8. 8.
    Vranic D, Saupe D (2001) 3D shape descriptor based on 3D fourier transform. In: Proceedings of the EURASIP conference on digital signal processing for multimedia communications and services (ECMCS 2001), Budapest, Hungary, pp 271–274Google Scholar
  9. 9.
    Australian Sign Language Dataset (1999) http://kdd.ics.uci.edu/databases/auslan/auslan.html
  10. 10.
    Bashir F, Khokhar A, Schonfeld D (2005) Automatic object trajectory-based motion recognition using Gaussian mixture models. In: IEEE international conference on multimedia and expo, pp 1532–1535Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2013

Authors and Affiliations

  1. 1.Chonnam National UniversityBuk-GuKorea
  2. 2.Chonbuk National UniversityDeokjin-guKorea
  3. 3.Chonnam National UniversityBuk-GuKorea

Personalised recommendations