摘要
运动特征的选择直接影响人体动作识别方法的识别效果。单一特征往往受到人体外观、环境、摄像机设置等因素的影响不同,其适用范围不同,识别效果也是有限的。在研究人体动作的表征与识别的基础上,充分考虑不同特征的优缺点,提出一种结合全局的剪影特征和局部的光流特征的混合特征,并用于人体动作识别。实验结果表明,该算法得到了理想的识别结果,对于Weizmann数据库中的动作可以达到100%的正确识别率。
The choice of the motion features affects the result of the human action recognition method directly. Many factors often influence the single feature differently, such as appearance of human body, environment and video camera. So the accu- racy of action recognition is limited. On the basis of studying the representation and recognition of human actions, and giving full consideration to the advantages and disadvantages of different features, this paper proposed a mixed feature which com- bined global silhouette feature and local optical flow feature. This combined representation was used for human action recogni- tion. The experimental results demonstrate that this algorithm can recognize human actions and achieve high recognition rates. This algorithm achieves 100% correct recognition rate for the human actions in the Weizmann database.
出处
《计算机应用研究》
CSCD
北大核心
2013年第2期601-604,共4页
Application Research of Computers
基金
国家自然科学基金青年基金资助项目(61103123)
关键词
动作识别
剪影特征
光流特征
留一法
action recognition
silhouette
optical tlows
leave one out
