Orthogonal Semi-Supervised Marginal Fisher Analysis

Abstract:

In recent years, a variety of manifold-based learning dimensionality reduction techniques have been proposed, which attempt to project the original data into a lower dimensional feature space by preserving the local neighborhood structure. Among them, marginal fisher analysis (MFA) achieved high performance for face recognition. However, the optimal basis vectors obtained by MFA are non-orthogonal and MFA usually deteriorates when labeled information is insufficient. In order to resolve these problems, we present a new method called orthogonal semi-supervised marginal fisher analysis (OSMFA), which not only extracts all the orthogonal discriminant vectors but also preserves the global structure of labeled and unlabeled samples to learn a better subspace for classification. Experimental results on ORL database demonstrate the effectiveness of the proposed algorithm.

• Periodical: Advanced Materials Research (Volumes 989-994)
• Pages: 2551-2554
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.989-994.2551
• Online since: July 2014
• Authors: Xing Zhu Liang*, Yu E Lin
• Keywords: Dimensionality Reduction, Face Recognition, Manifold-Based Learning, Orthogonal Semi-Supervised

References

[1] P Belhumeur, J Hespanha. et al. Eigenfaces vs. Fisherfaces: Recognition using class specific linear projection. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 19(1997), p.711–720.

DOI: 10.1109/34.598228

[2] X F He, S C Yan, Y Hu, et al. Face recognition using Laplacianfaces. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 27(2005), p.328–340.

DOI: 10.1109/tpami.2005.55

[3] D Xu, S Yan, D Tao , et al. Marginal Fisher Analysis and Its Variants for Human Gait Recognition and Content Based Image Retrieval. IEEE Transactions on Image processing, Vol. 16(2007), p.2811–2821.

DOI: 10.1109/tip.2007.906769

[4] D Cai, X F He, J W Han, Orthogonal laplacianfaces for face recognition. IEEE Transactions on Image Processing, Vol. 15(2006), p.3608–3614.

DOI: 10.1109/tip.2006.881945

[5] H F Hu. Orthogonal neighborhood preserving discriminant analysis for face recognition. Pattern Recognition, Vol. 41(2008), p.2045–(2054).

DOI: 10.1016/j.patcog.2007.10.029

[6] L Zhu, S N Zhu. Face recognition based on orthogonal discriminant locality preserving projections. Neurocomputing, Vol. 70(2007), p.1543–1546.

DOI: 10.1016/j.neucom.2006.12.004

[7] D Cai, X F He, J WHan. Semi-supervised discriminant analysis. Proceedings of the 11th IEEE International Conference on Computer Vision, ICCV07, Rio de Janeiro, Brazil, October, 2007, p.1–7.

DOI: 10.1109/iccv.2007.4408856

[8] H Huang, J W Li, J M Liu. Future Enhanced semi-supervised local Fisher discriminant analysis for face recognition. Generation Computer Systems, vol. 31(2012), p.244–253.

DOI: 10.1016/j.future.2010.11.005

[9] J G Wan, E Sung, W Y Yau. Incremental two-dimensional linear discriminant analysis with applications to facerecognition. Neurocomputing, vol. 33( 2010), p.314–322.