Type of Document	Thesis
Author	Zhang, Qiang
Author's Email Address	zhangqiang100@yahoo.com
URN	etd-04082005-133212
Title	Appearance-Based Classification and Recognition Using Spectral Histogram Representations and Hierarchical Learning for Oca
Degree	Master of Science
Department	Computer Science, Department of
Advisory Committee	Advisor Name Title Xiuwen Liu Committee Chair David Whalley Committee Member Kyle Gallivan Committee Member
Keywords	Object Recognition Computer Vision Feature Extraction
Date of Defense	2005-03-22
Availability	unrestricted
Abstract This thesis is composed of two parts. Part one is on Appearance-Based Classification and Recognition Using Spectral Histogram Representations. We present a unified method for appearance-based applications including texture classification, 2D object recognition, and 3D object recognition using spectral histogram representations. Based on a generative process, the representation is derived by partitioning the frequency domain into small disjoint regions and assuming independence among the regions. This gives rise to a set of filters and a representation consisting of marginal distributions of those filter responses. We provide generic evidence for its effectiveness in characterizing object appearance through statistical sampling and in classification by visualizing images in the spectral histogram space. We use a multilayer perceptron as the classifier and propose a filter selection algorithm by maximizing the performance over training samples. A distinct advantage of the representation is that it can be effectively used for different classification and recognition tasks. The claim is supported by experiments and comparisons in texture classification, face recognition, and appearance-based 3D object recognition. The marked improvement over existing methods justifies the effectiveness of the generative process and the derived spectral histogram representation. Part two is on Hierarchical Learning for Optimal Component Analysis. Optimization problems on manifolds such as Grassmann and Stiefel have been a subject of active research recently. However the learning process can be slow when the dimension of data is large. As a learning example on the Grassmann manifold, optimal component analysis (OCA) provides a general subspace formulation and a stochastic optimization algorithm is used to learn optimal bases. In this paper, we propose a technique called hierarchical learning that can reduce the learning time of OCA dramatically. Hierarchical learning decomposes the original optimization problem into several levels according to a specifically designed hierarchical organization and the dimension of the data is reduced at each level using a shrinkage matrix. The learning process starts from the lowest level with an arbitrary initial point. The following approach is then applied recursively: (i) optimize the recognition performance in the reduced space using the expanded optimal basis learned from the next lower level as an initial condition, and (ii) expand the optimal subspace to the bigger space in a pre-specied way. By applying this decomposition procedure recursively, a hierarchy of layers is formed. We show that the optimal performance obtained in the reduced space is maintained after the expansion. Therefore, the learning process of each level starts with a good initial point obtained from the next lower level. This speeds up the original algorithm significantly since the learning is performed mainly in reduced spaces and the computational complexity is reduced greatly at each iteration. The effectiveness of the hierarchical learning is illustrated on two popular datasets, where the computation time is reduced by a factor of about 30 compared to the original algorithm.
Files	Filename       Size       Approximate Download Time (Hours:Minutes:Seconds)     28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access    thesis_Qiang.pdf 940.72 Kb 00:04:21 00:02:14 00:01:57 00:00:58 00:00:05