Hand posture recognition with co-training

	Hand posture recognition with co-training
	Yikai Fang; Jian Cheng; Jinqiao Wang; Kongqiao Wang; Jing Liu; Hanqing Lu
	2008
会议名称	International Conference on Pattern Recognition
会议录名称	无
会议日期	December 8-11, 2008
会议地点	Tampa, Florida, USA
摘要	As an emerging human-computer interaction approachvision based hand interaction is more natural and efficient. Howeverin order to achieve high accuracy, most of the existing hand posture recognition methods need a large number of labeled samples which is expensive or unavailable in practice. In this paper, a co-training based method is proposed to recognize different hand postures with a small quantity of labeled data. Hand postures examples are represented with different features and disparate classifiers are trained simultaneously with labeled data. Then the semi-supervised learning treats each new posture as unlabeled data and updates the classifiers in a cotraining framework. Experiments show that the proposed method outperforms the traditional methods with much less labeled examples.
关键词	无
文献类型	会议论文
条目标识符	http://ir.ia.ac.cn/handle/173211/13451
专题	紫东太初大模型研究中心_图像与视频分析
通讯作者	Jian Cheng
推荐引用方式 GB/T 7714	Yikai Fang,Jian Cheng,Jinqiao Wang,et al. Hand posture recognition with co-training[C],2008.