• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.1
• /
• pp.25-34
• /
• 2010

Robust face recognition under various illumination environments is difficult to achieve. For robust face recognition with respect to illumination variations, illumination normalization of face images is usually applied as a preprocessing step. Most of previously proposed illumination normalization methods cannot handle cast shadows in face images effectively. In this paper, We propose a new face illumination normalization method based on the illumination-separated face identity texture subspace. Since the face identity texture subspace is constructed so as to be separated from the effects of illumination variations, the projection of face images into the subspace produces a good illumination-normalized face images. Through experiments, it is shown that the proposed face illumination normalization method can effectively eliminate cast shadows as well as attached shadows and achieves a good face illumination normalization.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.3 no.2
• /
• pp.5-16
• /
• 1999

We introduce a new projector for accelerating convergence of a symmetric eigenvalue problem Ax = x, and devise a power/Lanczos hybrid algorithm. Acceleration can be achieved by removing the hard-to-annihilate nonsolution eigencomponents corresponding to the widespread eigenvalues with modulus close to 1, by estimating them accurately using the Lanczos method. However, the additional Lanczos results can be obtained without expensive matrix-vector multiplications but a very small amount of extra work, by utilizing simple power-Lanczos interconversion algorithms suggested. Numerical experiments are given at the end.

• 제어로봇시스템학회:학술대회논문집
• /
• 1993.10a
• /
• pp.947-952
• /
• 1993

A simplified pole-placement design method is developed by analysing dynamic characteristics of the switching dynamics. Unlike the design procedure of conventional pole-placement, in the proposed method, overall state-space is directly decomposed into two invariant subspaces by the projection operator which is defined in the equivalent system, and then the closed-loop poles are assigned to each subspace independently. Hence, computations for state-feedback gain matrix are easy and simple.

• Speech Sciences
• /
• v.13 no.1
• /
• pp.117-128
• /
• 2006

The two-channel Generalized Sidelobe Canceller (GSC) scheme suffers from the presence of leakage signal in the reference channel. The leakage signal is caused by the dissimilar impulse responses between microphones, and different paths from speech source to microphones. Such leakage is detrimental to speech enhancement of the GSC since the desired reference signal becomes corrupted. In order to suppress the signal leakage, two matrix injection methods are proposed. In the first method, a simple gain compensation matrix is used. In the second, a projection matrix for reducing the error between the actual and the ideal primary and reference signals, is used. This paper describes the performance degradation resulting from leakage, and proposes effective methods to resolve the problem. Representative experiments were conducted to demonstrate the effectiveness of the proposed methods on recorded speech and noise in an actual automobile environment.

• ETRI Journal
• /
• v.38 no.6
• /
• pp.1095-1103
• /
• 2016

We present a method for transferring deformation weights of a human character to three-dimensional (3D) scanned clothes. First, clothing vertices are projected onto a character skin. Their deformation weights are determined from the barycentric coordinates of the projection points. For more complicated parts, such as shoulders and armpits, continuously moving planes are constructed and employed as projection reference planes. Clothing vertices on a plane are projected onto the intersection curve of the plane with a character skin to achieve a smooth weight transfer. The proposed method produces an initial deformation for physically based clothing simulations. We demonstrated the effectiveness of our method through several deformation results for 3D scanned clothes.

• Communications for Statistical Applications and Methods
• /
• v.26 no.4
• /
• pp.337-346
• /
• 2019

This paper discusses a method for obtaining nonnegative estimates for variance components in a random effects model. A variance component should be positive by definition. Nevertheless, estimates of variance components are sometimes given as negative values, which is not desirable. The proposed method is based on two basic ideas. One is the identification of the orthogonal vector subspaces according to factors and the other is to ascertain the projection in each orthogonal vector subspace. Hence, an observation vector can be denoted by the sum of projections. The method suggested here always produces nonnegative estimates using projections. Hartley's synthesis is used for the calculation of expected values of quadratic forms. It also discusses how to set up a residual model for each projection.

• Journal of the Chungcheong Mathematical Society
• /
• v.34 no.2
• /
• pp.181-188
• /
• 2021

Let X be a nondegenerate reduced closed subscheme in ℙⁿ. Assume that π_q : X → Y = π_q(X) ⊂ ℙ^n-1 is a generic projection from the center q ∈ Sec(X) \ X where Sec(X) = ℙⁿ. Let Z be the singular locus of the projection π_q(X) ⊂ ℙ^n-1. Suppose that I_X has the almost minimal presentation, which is of the form R(-3)^β_2,1 ⊕ R(-4) → R(-2)^β_1,1 → I_X → 0. In this paper, we prove the followings: (a) Z is either a linear space or a quadric hypersurface in a linear subspace; (b) $H^1({\mathcal{I}_X(k)})=H^1({\mathcal{I}_Y(k)})$ for all k ∈ ℤ; (c) reg(Y) ≤ max{reg(X), 4}; (d) Y is cut out by at most quartic hypersurfaces.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.7
• /
• pp.3194-3216
• /
• 2018

Slow Feature Discriminant Analysis (SFDA) is a supervised feature extraction method inspired by biological mechanism. In this paper, a novel method called Two Dimensional Slow Feature Discriminant Analysis via $L_{2,1}$ norm minimization ($2DSFDA-L_{2,1}$) is proposed. $2DSFDA-L_{2,1}$ integrates $L_{2,1}$ norm regularization and 2D statically uncorrelated constraint to extract discriminant feature. First, $L_{2,1}$ norm regularization can promote the projection matrix row-sparsity, which makes the feature selection and subspace learning simultaneously. Second, uncorrelated features of minimum redundancy are effective for classification. We define 2D statistically uncorrelated model that each row (or column) are independent. Third, we provide a feasible solution by transforming the proposed $L_{2,1}$ nonlinear model into a linear regression type. Additionally, $2DSFDA-L_{2,1}$ is extended to a bilateral projection version called $BSFDA-L_{2,1}$. The advantage of $BSFDA-L_{2,1}$ is that an image can be represented with much less coefficients. Experimental results on three face databases demonstrate that the proposed $2DSFDA-L_{2,1}/BSFDA-L_{2,1}$ can obtain competitive performance.

• Journal of KIISE:Databases
• /
• v.28 no.4
• /
• pp.568-576
• /
• 2001

Most of clustering algorithms data to degenerate rapidly on high dimensional spaces. Moreover, high dimensional data often contain a significant a significant of noise. which causes additional ineffectiveness of algorithms. Therefore it is necessary to develop algorithms adapted to the structure and characteristics of the high dimensional data. In this paper, we propose a clustering algorithms CLIP using the projection The CLIP is designed to overcome efficiency and/or effectiveness problems on high dimensional clustering and it is the is based on clustering on each one dimensional subspace but we use the incremental projection to recover high dimensional cluster and to reduce the computational cost significantly at time To evaluate the performance of CLIP we demonstrate is efficiency and effectiveness through a series of experiments on synthetic data sets.

• The Journal of the Acoustical Society of Korea
• /
• v.36 no.6
• /
• pp.401-406
• /
• 2017

This paper presents the performance assessment of several key algorithms conducted for amphibian species sound classification. Firstly, 9 target species including endangered species are defined and a database of their sounds is built. For performance assessment, three feature vectors such as MFCC (Mel Frequency Cepstral Coefficient), RCGCC (Robust Compressive Gammachirp filterbank Cepstral Coefficient), and SPCC (Subspace Projection Cepstral Coefficient), and three classifiers such as GMM(Gaussian Mixture Model), SVM(Support Vector Machine), DBN-DNN(Deep Belief Network - Deep Neural Network) are considered. In addition, i-vector based classification system which is widely used for speaker recognition, is used to assess for this task. Experimental results indicate that, SPCC-SVM achieved the best performance with 98.81 % while other methods also attained good performance with above 90 %.