Title :
A neurofuzzy approach for modeling lips movements
Author :
Bothe, H.H. ; Wieden, E.A.
Author_Institution :
Dept. of Electron., Tech. Univ. Berlin, Germany
Abstract :
The visual recognition is largely focussed on the speakers mouth region, especially on the lips. Since lips movements contain most of the visually perceptible information, this paper proposes the modeling of face movements with the help of related lips shapes only. The developed motion model consists of a multilayer neural network which selects the specific key-frames related to the single phonemes anf the surrounding next 3+3 neighbour phonemes. The feature vectors have been classified by means of a fuzzy c-mean algorithm
Keywords :
feedforward neural nets; fuzzy set theory; image recognition; speech recognition; face movements; fuzzy c-mean algorithm; lips movement modelling; multilayer neural network; neurofuzzy approach; visual speech recognition; Appraisal; Data mining; Lips; Loudspeakers; Mouth; Natural languages; Prototypes; Shape; Speech; Stress;
Conference_Titel :
Fuzzy Systems, 1994. IEEE World Congress on Computational Intelligence., Proceedings of the Third IEEE Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1896-X
DOI :
10.1109/FUZZY.1994.343679
