Estimating human age using bio-inspired features and the ranking method

@article{Spizhevoi2015EstimatingHA,
  title={Estimating human age using bio-inspired features and the ranking method},
  author={A. S. Spizhevoi and Alexander Vladimirovich Bovyrin},
  journal={Pattern Recognition and Image Analysis},
  year={2015},
  volume={25},
  pages={547-552}
}
This work is devoted to automatic estimation of human age from pictures of a face. An approach combining the use of bio-inspired features with the ranking method is suggested. For preliminary processing, methods of geometrical normalization and normalization of brightness are proposed. With a standard FG-NET set of data, an average absolute error of determination of 4.32 years, which is the best result that has been obtained from all the presently known methods, has been achieved. The obtained… CONTINUE READING

References

Publications referenced by this paper.
Showing 1-8 of 8 references

Mul  tiresolution gray  scale and rotation invariant texture classification with local binary patterns

  • M. Pietikainen Timo, T. Maenpaa
  • IEEE Trans . Pattern Anal . Mach . Intellig .
  • 2012

The way to modify an algorithm for classifying people ’ s age according to face image by using raking procedure

  • T. Lukina Rybintsev, V. Konushkin
  • IEEE Trans . Pattern Anal . Mach . Intellig .
  • 2001

Age synthe  sis and estimation via faces : a survey , ”

  • Guo Guodong, Guowang Mu, Yun Fu, T. S. Huang
  • IEEE Trans . Pat  tern Anal . Mach . Intellig .

Principal component analysis

  • L. J. Williams
  • The FG  NET Aging Database

Similar Papers

Loading similar papers…