In this paper , a framework of tied-mixture hidden Markov models ( TMHMM ) is proposed for vision-based SLR .
提出了基于混合元捆绑的隐马尔可夫模型（TMHMM）用于视觉手语识别。
tmhmm