Sept. 3, 2000 to Sept. 8, 2000
Sridha Sridharan , Queensland University of Technology
Vinod Chandran , Queensland University of Technology
Multimodal speech processing in which visual facial features are jointly processed with audio features is a rapidly advancing field. Lip movements and configurations provide useful information to improve speech and speaker recognition. However, the use of this visual information requires accurate and fast lip tracking algorithms. A new technique is outlined that is able to estimate the outer lip contour directly from a given lip intensity image via linear regression. An active shape model that is able to track speaker's lips without requiring time-consuming iterative energy minimization techniques can improve this estimate. Results of performance are presented against known tracking algorithms using the M2VTS database.
Sridha Sridharan, Vinod Chandran, "Initialized Eigenlip Estimator for Fast Lip Tracking Using Linear Regression", ICPR, 2000, Pattern Recognition, International Conference on, Pattern Recognition, International Conference on 2000, pp. 3182, doi:10.1109/ICPR.2000.903514