A Study of Face Motion Capture and Its Data Processing Technique Applied to the Speech Training of Hearing-impaired Children

Abstract

In this thesis the face motion capture technique is applied to the language rehabilitation training of hearing-impaired children with a presentation accordingly of an arrangement suited for the face marking feature points of hearing-impaired children in language rehabilitation training. Due to the fairly compact arrangement of human face, problems such as missing of marking points and noise interference can be crucial affecting factors on the training system. For such situation, the processing methods of the above two problems are conducted by means of data recovery measures and an application of adaptive Kalman filter to the suppression of noise propagation. Moreover, reasonable movement collection device and experiment parameters are selected and set up for use of data collection with the arrangement mentioned in the thesis. The results showed that this reasonably set arrangement of face marking feature points is highly efficient in the suppression of noise propagation.