Motion Energy Guided Multi-scale Heterogeneous Features for 3D Action Recognition

被引:0
作者
Liang, Chengwu [1 ]
Qi, Lin [1 ]
Guan, Ling [2 ]
机构
[1] Zhengzhou Univ, Sch Informat Engn, Zhengzhou, Henan, Peoples R China
[2] Ryerson Univ, Dept Elect & Comp Engn, Toronto, ON, Canada
来源
2017 IEEE VISUAL COMMUNICATIONS AND IMAGE PROCESSING (VCIP) | 2017年
关键词
Action recognition; sub-action segmentation; heterogeneous feature; Collaborative Representation; ENSEMBLE;
D O I
暂无
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
This paper is to address the problem of human action recognition in depth sequences. The actions with various speeds and shared sub-actions make the recognition challenging. A new feature set, consisting of two heterogeneous features are proposed to address this challenge. Specifically, we propose an adaptive normalized action motion energy based on the depth video. Guided by this multi-scale energy vector, depth sequence and skeleton pose sequence are divided respectively into two sets of subsequences with multiple scales (i.e., multi-scale sub-actions). Then in depth modality, based on the depth sub-sequence, Depth Motion Maps (DMMs) based Histogram Oriented Gradient (HOG) features are employed to capture the shape information and motion cues. In skeleton modality, based on the pose sub-sequence, pose dynamics using skeleton information are extracted. In order to obtain discriminative and compact representation, the Collaborative Representation (CR) learning scheme based classifier is adopted. Experiments on two datasets show the effectiveness of the proposed method.
引用
收藏
页数:4
相关论文
共 24 条
[1]   Human Activity Analysis: A Review [J].
Aggarwal, J. K. ;
Ryoo, M. S. .
ACM COMPUTING SURVEYS, 2011, 43 (03)
[2]  
[Anonymous], 2012, IEEE COMP SOC C COMP, DOI DOI 10.1109/CVPRW.2012.6239234
[3]  
[Anonymous], 2012, P ACM INT C MULT NAR, DOI DOI 10.1145/2393347.2396382
[4]  
Aydin R, 2014, IN C IND ENG ENG MAN, P1, DOI 10.1109/IEEM.2014.7058588
[5]  
Chen CC, 2013, INT J DISTRIB SENS N, DOI [10.1007/978-3-642-36373-3_1, 10.1155/2013/620248]
[6]   Representation Learning of Temporal Dynamics for Skeleton-Based Action Recognition [J].
Du, Yong ;
Fu, Yun ;
Wang, Liang .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2016, 25 (07) :3010-3022
[7]   Application on Integration Technology of Visualized Hierarchical Information [J].
Li, Weibo ;
He, Yang .
2010 THE 3RD INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND INDUSTRIAL APPLICATION (PACIIA2010), VOL I, 2010, :9-12
[8]   A Deep Structured Model with Radius-Margin Bound for 3D Human Activity Recognition [J].
Lin, Liang ;
Wang, Keze ;
Zuo, Wangmeng ;
Wang, Meng ;
Luo, Jiebo ;
Zhang, Lei .
INTERNATIONAL JOURNAL OF COMPUTER VISION, 2016, 118 (02) :256-273
[9]   Group Sparsity and Geometry Constrained Dictionary Learning for Action Recognition from Depth Maps [J].
Luo, Jiajia ;
Wang, Wei ;
Qi, Hairong .
2013 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2013, :1809-1816
[10]   HON4D: Histogram of Oriented 4D Normals for Activity Recognition from Depth Sequences [J].
Oreifej, Omar ;
Liu, Zicheng .
2013 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2013, :716-723