基于运动及形状原语的人体动作识别.pdf

第 12 卷 第 4 期 太赫兹科学与电子信息学报 Vo1．12，No．4 2014 年 8 月 Journal of Terahertz Science and Electronic Information Technology Aug．，2014 文章编号：2095-4980(2014)04-0572-07 基于运动及形状原语的人体动作识别 应 锐，王新宇，冯 辉，杨 涛，胡 波 (复旦大学 电子工程系，上海 200433) 摘 要 ：为了进行动作识别，需要对动作的运动与形状信息进行特征表示与建模。由于动作 可以由一系列原语即子动作构成，提出了一种新的基于运动特征和形状特征的原语提取及动作表 示方法。该算法首先从动作的运动特性出发，通过分割每一帧中的运动区域来提取运动特征，同 时利用梯度直方图提取形状特征。通过分层聚类的方式提取具有类内代表性及类间判决性的原语， 并利用混合高斯模型对原语进行特征表示。最后利用基于运动及形状原语的序列匹配进行动作识 别。实验表明，该算法在KTH和UCF 数据集上较现有方法取得了更高的识别正确率，分别为95.14% 和90.00% 。 关键词 ：运动；形状；原语；分层聚类；动作识别 中图分类号 ：TN911.7 ；TP391.41 文献标识码 ：A doi ：10.11805/TKYDA201404.0572 Human action recognition based on motion and shape primitives YING Rui，WANG Xin-yu，FENG Hui，YANG Tao，HU Bo (Department of Electronic Engineering，Fudan University，Shanghai 200433，China) Abstract： It is necessary to represent and model the motion and shape information of action in order to recognize the action. Considering the fact that action consists of a series of sub-actions namely primitives, a new algorithm of primitive extraction and action representation is proposed based on motion features and shape features. According to motion nature of action, motion feature is extracted through splitting the motion regions in each frame. Shape feature is also extracted via histogram of gradient. Then intra-class representative and inter-class discriminative primitives are extracted by hierarchical clustering and represented by Gaussian Mixture Model. Finally, action recognition is executed based on motion and shape primitives via sequence matching. The