Title :
Low-dimensional processing of still and moving images
Author_Institution :
Dept. of Math., Colorado State Univ., Fort Collins, CO, USA
Abstract :
The Karhunen-Loeve procedure for the processing of still and moving images is considered. The formulation is presented in terms of the singular value decomposition, an approach which makes it possible to treat moderate sized ensembles of high-resolution images or large ensembles of low-resolution images. Applications of the technique to several test problems, including data generated by numerical and laboratory simulations of fluid motion and digital images of human faces and animated lips, are reviewed
Keywords :
face recognition; fluid dynamics; image processing; image sequences; Karhunen-Loeve procedure; SVD; animated lips; digital images; fluid motion; high-resolution images; human faces; low-dimensional processing; low-resolution images; moving images; singular value decomposition; still images; Character generation; Digital images; Face; Facial animation; Humans; Laboratories; Mathematics; Motion analysis; Test pattern generators; Testing;
Conference_Titel :
Signals, Systems and Computers, 1992. 1992 Conference Record of The Twenty-Sixth Asilomar Conference on
Conference_Location :
Pacific Grove, CA
Print_ISBN :
0-8186-3160-0
DOI :
10.1109/ACSSC.1992.269142
