Advanced search
1 file | 263.56 KB

Real-time multi-colourspace hand segmentation

Author
Organization
Abstract
This paper proposes an accurate real-time hand tracking and segmentation algorithm. A particle filter tracks the hands in time, based on colour and motion cues. This filter is able to automatically recover from failures and does not need an initialization phase. The algorithm is proven to be robust against lighting changes, and can be used in unconstrained environments. Hand segmentation is based on a Gaussian Mixture Model and refined using a combination of spatial information. Cues from both HSV and RGB colour space are used to increase robustness.
Keywords
hand tracking, hand segmentation, hand detection

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 263.56 KB

Citation

Please use this url to cite or link to this publication:

Chicago
Spruyt, Vincent, Alessandro Ledda, and Stig Geerts. 2010. “Real-time Multi-colourspace Hand Segmentation.” In IEEE International Conference on Image Processing ICIP, 3117–3120. New York, NY, USA: IEEE.
APA
Spruyt, V., Ledda, A., & Geerts, S. (2010). Real-time multi-colourspace hand segmentation. IEEE International Conference on Image Processing ICIP (pp. 3117–3120). Presented at the 2010 IEEE 17th International conference on Image Processing (ICIP 2010), New York, NY, USA: IEEE.
Vancouver
1.
Spruyt V, Ledda A, Geerts S. Real-time multi-colourspace hand segmentation. IEEE International Conference on Image Processing ICIP. New York, NY, USA: IEEE; 2010. p. 3117–20.
MLA
Spruyt, Vincent, Alessandro Ledda, and Stig Geerts. “Real-time Multi-colourspace Hand Segmentation.” IEEE International Conference on Image Processing ICIP. New York, NY, USA: IEEE, 2010. 3117–3120. Print.
@inproceedings{1156821,
  abstract     = {This paper proposes an accurate real-time hand tracking and segmentation algorithm. A particle filter tracks the hands in time, based on colour and motion cues. This filter is able to automatically recover from failures and does not need an initialization phase. The algorithm is proven to be robust against lighting changes, and can be used in unconstrained environments. Hand segmentation is based on a Gaussian Mixture Model and refined using a combination of spatial information. Cues from both HSV and RGB colour space are used to increase robustness.},
  author       = {Spruyt, Vincent and Ledda, Alessandro and Geerts, Stig},
  booktitle    = {IEEE International Conference on Image Processing ICIP},
  isbn         = {9781424479948},
  issn         = {1522-4880},
  keyword      = {hand tracking,hand segmentation,hand detection},
  language     = {eng},
  location     = {Hong Kong, PR China},
  pages        = {3117--3120},
  publisher    = {IEEE},
  title        = {Real-time multi-colourspace hand segmentation},
  url          = {http://dx.doi.org/10.1109/ICIP.2010.5653220},
  year         = {2010},
}

Altmetric
View in Altmetric
Web of Science
Times cited: