Computer Engineering and Applications ›› 2017, Vol. 53 ›› Issue (19): 211-215.DOI: 10.3778/j.issn.1002-8331.1604-0136

Previous Articles     Next Articles

Spatio-temporal pyramid for action recognition based on depth sequences

ZOU Xiangyang1,2, HOU Yunjiang1   

  1. 1.Air Force Airborne Academy, Guilin, Guangxi 541003, China
    2.College of Electronic Engineering and Automation, Guilin University of Electronic Technology, Guilin, Guangxi 541004, China
  • Online:2017-10-01 Published:2017-10-13

基于深度序列的时空金字塔的动作识别

邹向阳1,2,侯云江1   

  1. 1.空军空降兵学院,广西 桂林 541003
    2.桂林电子科技大学 电子工程与自动化学院,广西 桂林 541004

Abstract: An efficient human action recognition method is put forward. Three views of depth sequences are transformed into Depth Motion Outline Sequence(DMOS) by using the method of interframe differentiation. Then a spatio-temporal pyramid is proposed to subdivide the DMOS on temporal and spatial level. A feature fusion scheme is presented to concatenate the Histograms of Oriented Gradients(HOG) features which have extracted from the subdivided DMOS. Finally linear SVM to classification is used. Through using the MSR Action 3D data sets, this method is evaluated with different parameters of spatial-temporal pyramid. Experimental results show that this method has higher recognition rate than the similar algorithm.

Key words: action recognition, Depth Motion Outline Sequence(DMOS), spatio-temporal pyramid, Histogram of Oriented Gradient(HOG), linear SVM classifier

摘要: 提出一种高效的人体动作识别方法。通过帧间差分法将深度序列的三视图转化为深度运动轮廓序列(DMOS),然后利用时空金字塔对DMOS进行时间维和空间维细分,将细分后得到的空间网格的局部方向梯度直方图(HOG)进行特征融合,并使用线性SVM分类。最后采用MSR Action 3D数据集对提出的算法在不同时空金字塔参数下的识别率和处理速度进行了评估,结果表明该方法在同类算法中具有更高的识别率。

关键词: 动作识别, 深度运动轮廓序列, 时空金字塔, HOG特征, 线性SVM分类器