Data Fusion for Geometrical and Pixel Based Lip Feature

Author

Wang Mengjun ; Li Gang

Author_Institution

Sch. of Inf. Eng., HeBei Univ. of Technol., Tianjin, China

fYear

2010

fDate

28-29 Oct. 2010

Firstpage

301

Lastpage

304

Abstract

Lipreading is applied to synthesize speech for the speech-impaired people. To get a higher recognition result, data fusion with weighting coefficients at feature level is used to integrate the lip information from different kinds of lip features. Experiments are carried out based on HMM with different states and Gaussian mixture component in a small database for speaker-dependent case. Experiment results showed that the integrated discriminate vector after feature fusion obtains the information from the Geometrical feature vector of lip region and the DCT coefficients of lip´ ROI. With best weighting coefficients m: n=1.5:1, the recognition rate are improved by as much as 5.02% and 8.37%, respectively.

Keywords

Gaussian processes; discrete cosine transforms; feature extraction; handicapped aids; hidden Markov models; sensor fusion; speech synthesis; DCT coefficients; Gaussian mixture component; HMM; data fusion; geometrical based lip feature; integrated discriminate vector; lipreading; pixel based lip feature; speech synthesis; speech-impaired people; Discrete cosine transforms; Feature extraction; Hidden Markov models; Image sequences; Pixel; Speech; Visualization; Hidden Markov Model; data fusion; geometrical based feature vector; pixel based feature vector; weighting combination;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligence Information Processing and Trusted Computing (IPTC), 2010 International Symposium on

Conference_Location

Huanggang

Print_ISBN

978-1-4244-8148-4

Electronic_ISBN

978-0-7695-4196-9

Type

conf

DOI

10.1109/IPTC.2010.35

Filename

5663130