Sequential Lip Region Segmentation Using Fuzzy Clustering with Spatial and Temporal Information
For many visual speech recognition and visual speaker authentication systems, lip region extraction is of vital important. In order to segment the lip region accurately and robustly from a lip sequence, a new fuzzy-clustering based algorithm is proposed. In the proposed method, a new dissimilarity measure is introduced to take all the color, spatial and temporal information into consideration. An iterative optimization method is employed to derive the optimal lip region membership map and the final segmentation result. From the experimental results, it is observed that the proposed algorithm can provide superior results compared with other traditional methods.
KeywordsTemporal Information Fuzzy Cluster Final Segmentation Result Iterative Optimization Method Image Ambiguity
Unable to display preview. Download preview PDF.
- 1.Petajan, E.D.: Automatic lip-reading to enhance speech recognition. In: Proc. of IEEE Conference on Computer Vision and Pattern Recognition, pp. 40–47 (1985)Google Scholar
- 2.Zhang, Y., Levinson, S., Huang, T.: Speaker independent audio-visual speech recognition. In: Proc. of IEEE Int. Conf. on Multimedia and Expo. New York, vol. 2, pp. 1073–76 (2000)Google Scholar
- 3.Sadeghi, M., Kittler, J., Messer, K.: Modeling and segmentation of lip area in face images. In: Proc. of IEEE Vision, Image and Signal Processing, vol. 149, pp. 179–184 (2002)Google Scholar
- 4.Eveno, N., Caplier, A., Coulon, P.Y.: New color transformation for lips segmentation. In: Proceedings of IEEE Fourth Workshop on Multimedia Signal Processing, Cannes, France, pp. 3–8 (October 2001)Google Scholar
- 5.Rohani, R., lizadeh, S.A., Sobhanmanesh, F., Boostani, R.: Lip Segmentation in Color Images. In: Proc. of International Conference on Innovations in Information Technology, pp. 747–759 (2008)Google Scholar
- 6.Wang, S.L., Leung, S.H., Lau, W.H.: Lip segmentation by fuzzy clustering incorporating with shape function. In: Proc. of IEEE ICAASP 2002, Orlando, vol. 1, pp. 1077–1080 (2002)Google Scholar