Skip to main content
SpringerLink
Log in

Online nonparametric discriminant analysis for incremental subspace learning and recognition

  • Theoretical Advances
  • Published:
Pattern Analysis and Applications Aims and scope Submit manuscript

Abstract

This paper presents a novel approach for online subspace learning based on an incremental version of the nonparametric discriminant analysis (NDA). For many real-world applications (like the study of visual processes, for instance) it is impossible to know beforehand the number of total classes or the exact number of instances per class. This motivated us to propose a new algorithm, in which new samples can be added asynchronously, at different time stamps, as soon as they become available. The proposed technique for NDA-eigenspace representation has been used in pattern recognition applications, where classification of data has been performed based on the nearest neighbor rule. Extensive experiments have been carried out both in terms of classification accuracy and execution time. On the one hand, the results show that the Incremental NDA converges towards the classical NDA at the end of the learning process and furthermore. On the other hand, Incremental NDA is suitable to update a large knowledge representation eigenspace in real-time. Finally, the use of our method on a real-world application is presented.

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

Similar content being viewed by others

Notes

  1. In order to overcome the singularity problem, a PCA has been applied beforehand.

  2. For this experiment we selected 85 classes from the total of 116, because these classes correspond to people who took part in both photographic sessions. We also removed those images showing occlusions.

  3. The values in the ‘face size’ column represent the actual values used in our experiments obtained after downsampling and cropping the face area from the original images.

  4. For this experiment, in order to have more data, we considered also the images affected by occlusions.

  5. In the current study, we put the accent in having a reasonable number of classes with many instances rather having an excessive number of classes with very few instances.

References

  1. Artač M, Jogan M, Leonardis A (2002) Incremental PCA for on-line visual learning and recognition. In: Proceedings of the 16th international conference on pattern recognition, vol 3. Québec, Canada, pp 781–784

  2. Blake C, Merz C (1998) UCI repository of machine learning databases. http://www.ics.uci.edu/ mlearn/MLRepositor. html

  3. Bressan M, Vitrià J (2003) Nonparametric discriminant analysis and nearest neighbor classification. Pattern Recognit Lett 24(15):2743–2749

    Article  Google Scholar 

  4. Chandrasekaran S, Manjumath BS, Wang YF, Winkler J, Zhang H (1997) An eigenspace update algorithm for image analysis. Graph Models Image Process 59(5):321–332

    Article  Google Scholar 

  5. Duda RO, Hart PE, Stork DG (2001) Pattern classification, 2nd edn. Wiley, New York

    MATH  Google Scholar 

  6. Etemad K, Chellapa R (1997) Discriminant analysis for recognition of human face images. J Opt Soc Am 14(8):1724–1733

    Article  Google Scholar 

  7. Fidler S, Skočcaj D, Leonardis A (2006) Combining reconstructive and discriminative subspace methods for robust classification and regression by subsampling. IEEE Trans Pattern Anal Mach Intell 28(3):337–350

    Article  Google Scholar 

  8. Fisher RA (1938) The statistical utilization of multiple measurements. Ann Eugen 8:376–386

    Google Scholar 

  9. Fukunaga K (1990) Introduction to statistical pattern recognition, 2nd edn. Academic Press, Boston

    MATH  Google Scholar 

  10. Hall P, Marshall D, Martin R (1998) Incremental eigenanalysis for classification. In: Proceedings of British machine vision conference, vol 1. Southampton, UK, pp 286–295

  11. Kotropoulos C, Tefas A, Pitas I (2000) Frontal face authentication using morphological elastic graph matching. IEEE Trans Pattern Anal Mach Intell 9(4):555–560

    Google Scholar 

  12. Li Z, Liu W, Lin D, Tang X (2005) Nonparametric subspace analysis for face recognition. In: Proceedings of 2005 IEEE conference on computer vision and pattern recognition, vol 2. San Diego, USA, pp 961–966

  13. Martinez AM, Kak AC (2001) PCA versus LDA. IEEE Trans Pattern Anal Mach Intell 23(2):228–233

    Article  Google Scholar 

  14. Martinez AM, Benavente R (1998) The AR face database. Computer Vision Center. Technical report #24

  15. Masip D, Vitrià J (2005) Boosted linear projections for discriminant analysis. In: Vitrià et al (eds) Frontiers in artificial intelligence and applications series. Recent advances in artificial intelligence: Research and Development. IOS Press, Amsterdam, pp 45–52

  16. Samaria F, Harter A (1994) Parameterisation of a stochastic model for human face identification. In: Proceedings of second IEEE workshop on applications of computer vision, Sarasota

  17. Oja E (1983) Subspace methods of pattern recognition. Research Studies Press, Letchworth

    Google Scholar 

  18. Pang S, Ozawa S, Kasabov N (2005) Chunk incremental LDA computing on data streams. In: Wang et al (eds) International symposium on neural networks, vol LNCS 3497. Chongqing, China, pp 51–56

  19. Skočaj D, Uray M, Leonardis A, Bischof H (2006) Why to combine reconstructive and discriminative information for incremental subspace learning. In: Chum et al (eds) Proceedings of computer vision winter workshop, Telč, Czech Republic, pp N/A

  20. Swets DL, Weng JJ (1996) Using discriminant eigenfeatures for image retrieval. IEEE Trans Pattern Anal Machine Intell 18(8):831–836

    Article  Google Scholar 

  21. Graham DB, Allinson NM (1998) Characterizing virtual eigensignatures for general purpose face recognition. In: Wechsler et al (eds) Face recognition: from theory to applications, NATO ASI series F. Computer and systems sciences, vol 163, pp 446–456

  22. Viola P, Jones MJ (2004) Robust real-time face detection. Int J Comp Vis 57:137–154

    Article  Google Scholar 

  23. Weng JJ (1996) Crescepton and SHOSLIF: towards comprehensive visual learning. In: Nayar et al (eds) Early visual learning. Oxford University Press, Oxford, pp 183–214

  24. Weng J, Zhang Y, Hwang W-S (2003) Candid covariance-free incremental principal component analysis. IEEE Trans Pattern Recognit 25(8):1034–1040

    Article  Google Scholar 

  25. Zhou SK, Chellappa R (2004) Multiple-exemplar discriminant analysis for face recognition. In: Proceedings of IEEE conference on pattern recognition, Cambridge, UK, pp N/A

Download references

Acknowledgments

This work is supported by MCYT Grant TIN2006-15308-C02, Ministerio de Educación y Ciencia, Spain. Bogdan Raducanu is supported by the Ramon y Cajal research program, Ministerio de Educación y Ciencia, Spain.

Author information

Authors and Affiliations

  1. Computer Vision Center, Edifici “O”, Campus UAB, 08193, Bellaterra, Barcelona, Spain

    B. Raducanu & J. Vitrià

  2. Department of Applied Mathematics and Analysis, University of Barcelona, 08003, Barcelona, Spain

    J. Vitrià

Authors
  1. B. Raducanu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. J. Vitrià
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to B. Raducanu.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Raducanu, B., Vitrià, J. Online nonparametric discriminant analysis for incremental subspace learning and recognition. Pattern Anal Applic 11, 259–268 (2008). https://doi.org/10.1007/s10044-008-0131-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10044-008-0131-0

Keywords

Search

Navigation