Abstract
In the gait recognition, the dependency to the direction of walking is very serious problem. To reduce this dependency, we propose a view synthesis method based on the planar homography. However, even though the synthesized gait is used, shape information is not enough to recognize individual. Thus, in this paper, we use shape sequence descriptor for recognition, which describes shape information and variation according to motion at a same time. Our experiments show that the proposed method efficiently reduces the dependency to directional variations of gait.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Lee, L., Grimson, W.E.L.: Gait Appearance for Recognition. In: Proc. ECCV Workshop, Biometric Authentication, pp. 143–154 (2002)
BenAbdelkader, C., Cutler, R.: View-invariant Estimation of Height and Stride for Gait Recognition. In: Proc. ECCV Workshop on Biometric Authentication, pp. 155–167 (2002)
RoyChowdhury, A.K., Kale, A., Chellappa, R.: Video Synthesis of Arbitrary Views for Approximately Planar Scenes. In: Proc. IEEE Conf. on Acoustics, Speech and Signal Processing, vol. 3, pp. 497–500 (2003)
Hartley, R., Zisserman, A.: Multiple View Geometry in Computer Vision. Cambridge University Press (2000)
Choi, M.S.: An Efficient Representation and Matching of a Moving Object in Silhouette, Ph.D. Thesis, Hanyang University (2004)
Bober, M.: Mpeg-7 Visual Shape Descriptors. IEEE Trans. Circuits System for Video Technology 11, 716–719 (2001)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Jeong, S., Lee, Y., Lee, KW., Cho, J. (2011). Robust Gait Recognition Using Planar Homography and Shape Sequence Descriptor. In: Kim, Th., et al. Grid and Distributed Computing. GDC 2011. Communications in Computer and Information Science, vol 261. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-27180-9_59
Download citation
DOI: https://doi.org/10.1007/978-3-642-27180-9_59
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-27179-3
Online ISBN: 978-3-642-27180-9
eBook Packages: Computer ScienceComputer Science (R0)