W. Brent Seales, Cheng J. Yuan, Wei Hu, and Mattew d. Cutts. Content analysis of compressed video. Technical Report 2 65-96, University of Kentucky Computer Science Dept., Lexington, Kentucky, August 1996. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Exploiting Image Indexing Techniques in DCT Domain - Ngo, Pong, Chin (1998) (5 citations) (Correct)
....For instance, the texture feature is formed by computing the statistical measures of the DCT coefficients. To reduce the dimensionality of the feature space, one can employ the Fisher Discriminant techniques to maximize the separability among the known texture classes. Similarly, Seales et al. [10] employ principle component analysis to obtain eigenvectors from DCT coefficients. Since DCT is linear and orthogonal, the distance in eigenspace is preserved. The authors project the DCT space to the first few principle axes and perform object recognition directly in the compressed domain. These ....
W. Brent Seales, C. J. Yuan & W. Hu.: Content Analysis of Compressed Video. Technical Report 265-96, University of Kentucky, 1996.
Wedding the Wavelet Transform and Multivariate Data Analysis - Murtagh (1996) (Correct)
....these transforms are commonly used for data cleaning or filtering in image processing. As opposed to data smoothing, they are used to achieve much the same objective of good data approximation, but with fine features of the data respected. PCA, DCT, and FT have the property of energy packing (Seales et al. 1996): most of the energy (second order moment, or frequency squared) of the input vector is packed into the first few values of the output vector. Thus, one can roughly approximate, or even eliminate, all but the most important values and still preserve most of the input energy. 10 The wavelet ....
....we come to the issue of major emphasis in this article: how much variation in data analysis performance can be expected as a result of the errors introduced into the data set from lossy compression Clearly an answer depends both on the transform used and on the effectiveness of denoising. Seales et al. 1996) use the very widely supported JPEG (Joint Photographic Experts Group) image compression standard which is based on DCT, and then tests PCA results on the compressed data compared to PCA of the original data. The aim is to construct principal component images of video sequences of human faces ....
SEALES, W.B., YUAN, C.J., HU W., and CUTTS, M.D. (1996), "Content Analysis of Compressed Video", University of Kentucky Computer Science Department Technical Report No. 2 65--96 August 1996 (28 pp.).
Image Processing through Multiscale Analysis and.. - Murtagh And Starck (2000) (2 citations) (Correct)
....we also include the discrete cosine transform (DCT) which decomposes the data into an orthogonal basis of cosine functions; and the Fourier transform (FT) which uses a basis of sine and cosine functions, each at different frequencies. PCA, DCT, and FT have the property of energy packing (Seales et al. 1996): most of the energy (second order moment) of the input vector is packed into the first few values of the output vector. Thus, one can roughly approximate, or even eliminate, all but the most important values and still preserve most of the input energy. The wavelet transform (WT) whether ....
Seales, W.B., Yuan, C.J., Hu W. and Cutts, M.D. (1996) Content analysis of compressed video, University of Kentucky Computer Science Department Technical Report No. 2, 28 pp.
Motion Tracking In Active Vision System - Yurong Li University (Correct)
No context found.
W. Brent Seales, Cheng J. Yuan, Wei Hu, and Mattew d. Cutts. Content analysis of compressed video. Technical Report 2 65-96, University of Kentucky Computer Science Dept., Lexington, Kentucky, August 1996.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC