A support system for generating sign language animation using thermal image processing
Taro Asada, Yasunari Yoshitom · 2010
We have proposed an approach aimed at sign language animation by skin region detection on an infrared image. In the system implemented the proposed method, a 3D CG model corresponding to a characteristic posture in sign language is made automatically by pattern recognition on a thermal image, and then a person's hand in the CG model is set. The hand part is made manually beforehand. If necessary, the model can be replaced manually by a more appropriate model corresponding to training key frames and/or the model can be refined manually. In our experiments, three persons experienced in using sign language recognized the Japanese sign language of 70 words expressed as animation with 94.3% accuracy. Then, we improved the system to correct the position and the direction of hand of the model automatically generated, using fuzzy algorithm and simulated annealing.