Lipreading from color video

被引：79

作者：

Chiou, GI

Hwang, JN

机构：

[1] Information Processing Laboratory, Department of Electrical Engineering, University of Washington, Seattle

来源：

IEEE TRANSACTIONS ON IMAGE PROCESSING | 1997年 / 6卷 / 08期

关键词：

active contour model; hidden Markov model; Karhunen-Loeve transform; lipreading; snake; visual phoneme;

D O I：

10.1109/83.605417

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We have designed and implemented a lipreading system that recognizes isolated words using only color video of human lips (without acoustic data). The system performs video recognition using ''snakes'' to extract visual features of geometric space, Karhunen-Loeve transform (KLT) to extract principal components in the color eigenspace, and hidden Markov models (HMM's) to recognize the combined visual features sequences. With the visual information alone, we were able to achieve 94% accuracy for ten isolated words.

引用

页码：1192 / 1195

页数：4

共 19 条

[1]

[Anonymous], 1992, 3 BRIT MACH VIS C 19

[2]

[Anonymous], 1995, P INT WORKSH AUT FAC

[3]

BREGLER C, 1994, INT CONF ACOUST SPEE, P669, DOI 10.1109/ICASSP.1994.389567

[4] A NEURAL-NETWORK-BASED STOCHASTIC ACTIVE CONTOUR MODEL (NNS-SNAKE) FOR CONTOUR FINDING OF DISTINCT FEATURES [J].