Visual speech recognition method using translation, scale and rotation invariant features

Yau, W, Kumar, D and Poosapadi Arjunan, S 2006, 'Visual speech recognition method using translation, scale and rotation invariant features', in M. Piccardi (ed.) IEEE International Conference on Video and Signal Based Surveillance 2006, Sydney, Australia, November 2006.


Document type: Conference Paper
Collection: Conference Papers

Title Visual speech recognition method using translation, scale and rotation invariant features
Author(s) Yau, W
Kumar, D
Poosapadi Arjunan, S
Year 2006
Conference name IEEE International Conference on Video and Signal Based Surveillance
Conference location Sydney, Australia
Conference dates November 2006
Proceedings title IEEE International Conference on Video and Signal Based Surveillance 2006
Editor(s) M. Piccardi
Publisher IEEE
Place of publication Sydney, Australia
Abstract This paper reports on a visual speech recognition method that is invariant to translation, rotation and scale. Dynamic features representing the mouth motion is extracted from the video data by using a motion segmentation technique termed as motion history image (MHI). MHI is generated by applying accumulative image differencing technique on the sequence of mouth images. Invariant features are derived from the MHI using feature extraction algorithm that combines Discrete Stationary Wavelet Transform (SWT) and moments. A 2-D SWT at level one is applied to decompose MHI to produce one approximate and three detail sub images. The feature descriptors consist of three moments (geometric moments, Hu moments and Zernike moments) computed from the SWT approximate image. The moments features are normalized to achieve the invariance properties. Artificial neural network (ANN) with back propagation learning algorithm is used to classify the moments features. Initial experiments were conducted to test the sensitivity of the proposed approach to rotation, translation and scale of the mouth images and obtained promising results.
Subjects Biomedical Engineering not elsewhere classified
DOI - identifier 10.1109/AVSS.2006.118
Copyright notice © 2006 IEEE
Versions
Version Filter Type
Citation counts: Scopus Citation Count Cited 5 times in Scopus Article | Citations
Altmetric details:
Access Statistics: 170 Abstract Views  -  Detailed Statistics
Created: Wed, 08 Apr 2009, 09:42:32 EST by Catalyst Administrator
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us