This Article 
 Bibliographic References 
 Add to: 
Clustered Blockwise PCA for Representing Visual Data
October 2005 (vol. 27 no. 10)
pp. 1675-1679
Principal Component Analysis (PCA) is extensively used in computer vision and image processing. Since it provides the optimal linear subspace in a least-square sense, it has been used for dimensionality reduction and subspace analysis in various domains. However, its scalability is very limited because of its inherent computational complexity. We introduce a new framework for applying PCA to visual data which takes advantage of the spatio-temporal correlation and localized frequency variations that are typically found in such data. Instead of applying PCA to the whole volume of data (complete set of images), we partition the volume into a set of blocks and apply PCA to each block. Then, we group the subspaces corresponding to the blocks and merge them together. As a result, we not only achieve greater efficiency in the resulting representation of the visual data, but also successfully scale PCA to handle large data sets. We present a thorough analysis of the computational complexity and storage benefits of our approach. We apply our algorithm to several types of videos. We show that, in addition to its storage and speed benefits, the algorithm results in a useful representation of the visual data.

[1] S. Avidan, “EigenSegments: A Spatio-Temporal Decomposition of an Ensemble of Images,” Proc. European Conf. Computer Vision, vol. 3, pp. 747-758, 2002.
[2] C.M. Bishop, “Bayesian PCA,” Advances in Neural Information Processing Systems, vol. 11, pp. 382-388, 1999.
[3] J.R. Bunch and C.P. Nielsen, “Updating the Singular Value Decomposition,” Numerische Mathematik, vol. 31, pp. 111-129, 1978.
[4] S. Chandrasekaran, B.S. Manjunath, Y.F. Wang, J. Winkler, and H. Zhang, “An Eigenspace Update Algorithm for Image Analysis,” Graphical Models and Image Processing, vol. 59, no. 5, pp. 321-332, 1997.
[5] F.R.K. Chung, Spectral Graph Theory. Am. Math. Soc., 1997.
[6] P. Debevec, T. Hawkins, C. Tchou, H.-P. Duiker, W. Sarokin, and M. Sagar, “Acquiring the Reflectance Field of a Human Face,” Proc. ACM SIGGRAPH 2000, pp. 145-156, July 2000.
[7] K. Fukunaga, Introduction to Statistical Pattern Recognition. Academic Press, 1990.
[8] G.H. Golub and C.F. Van Loan, Matrix Computation. Johns Hopkins Univ. Press, 1996.
[9] M. Gu and S.C. Eisenstat, “A Stable and Efficient Algorithm for the Rank-One Modification of the Symmetric Eigenproblem,” SIAM J. Matrix Analysis and Application, vol. 15, no. 4, pp. 1266-1276, 1994.
[10] P. Hall, D. Marshall, and R. Martin, “Merging and Splitting Eigenspace Models,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 22, no. 9, pp. 1042-1049, Sept. 2000.
[11] P. Hallinan, “A Low-Dimensional Representation of Human Faces for Arbitrary Lighting Conditions,” Proc. Computer Vision and Pattern Recognition, pp. 995-999, 1994.
[12] H. Hotelling, “Analysis of a Complex Of Statistical Variables into Principal Components,” J. Educational Psychology, vol. 24, pp. 417-441, 1933.
[13] T. Jebara, K. Russell, and A. Pentland, “Mixtures of Eigenfeatures for Real-Time Structure from Texture,” Proc. Int'l Conf. Computer Vision, pp. 128-135, 1998.
[14] I.T. Jolliffe, Principal Component Analysis. New York: Springer-Verlag, 1986.
[15] A. Leonardis, H. Bischof, and J. Maver, “Multiple Eigenspaces,” Pattern Recognition, vol. 35, pp. 2613-2627, 2002.
[16] A. Levin and A. Shashua, “Principal Component Analysis Over Continuous Subspaces and Intersection of Half-Spaces,” Proc. European Conf. Computer Vision, vol. 3, pp. 635-650, 2002.
[17] Z. Lin, T.-T. Wong, and H.-Y. Shum, “Relighting with the Reflected Irradiance Field: Representation, Sampling and Reconstruction,” Int'l J. Computer Vision, vol. 49, nos. 2-3, pp. 229-246, 2002.
[18] W. Matusik, H. Pfister, A. Ngan, P. Beardsley, R. Ziegler, and L. McMillan, “Image-Based 3D Photography Using Opacity Hulls,” ACM Trans. Graphics, vol. 21, no. 3, pp. 427-437, 2002.
[19] B. Moghaddam and A. Pentland, “Probalistic Visual Learning for Object Detection,” Proc. Int'l Conf. Computer Vision, pp. 786-793, 1995.
[20] H. Murakami and B.V.K.V. Kumar, “Efficient, Numerically Stabilized Rank-One Eigenstructure Updating,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 4, no. 5, pp. 511-515, 1982.
[21] H. Murase and M. Lindenbaum, “Spatial Temporal Adaptive Method for Partial Eigenstructure Decomposition of Large Image Matrices,” IEEE Trans. Image Processing, vol. 4, no. 5, pp. 620-629, 1995.
[22] K. Pearson, “On Lines and Planes of Closest Fit to Systems of Points in Space,” The London, Edinburgh, and Dublin Philosophical Magazine and J. Science, six series 2, pp. 559-572, 1901.
[23] A. Shashua, “On Photometric Issues in 3D Visual Recognition from a Single 2D Image,” Int'l J. Computer Vision, vol. 21, no. 1, pp. 99-122, 1997.
[24] E.P. Simoncelli and B.A. Olshausen, “Natural Image Statistics and Neural Representation,” Ann. Rev. Neuroscience, vol. 24 pp. 1193-1216, 2001.
[25] L. Sirovich, “Turbulence and the Dynamics of Coherent Structures,” Quarterly Applied Math., vol. 45, pp. 561-571, 1987.
[26] G.W. Stewart and J.-G. Sun, Matrix Perturbation Theory. Academic Press, 1990.
[27] L. Zhao and Y.-H. Yang, “Mosaic Image Method: A Local and Global Method,” Pattern Recognition, vol. 32, no. 8, pp. 1421-1434, 1997.

Index Terms:
Index Terms- Principal component analysis, singular value decomposition, eigenvalues and eigenvectors, natural image statistics, clustering, region growing/partitioning.
Ko Nishino, Shree K. Nayar, Tony Jebara, "Clustered Blockwise PCA for Representing Visual Data," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 27, no. 10, pp. 1675-1679, Oct. 2005, doi:10.1109/TPAMI.2005.193
Usage of this product signifies your acceptance of the Terms of Use.