基于深度信息的人体动作识别研究综述

被引:30
作者
陈万军
张二虎
机构
[1] 西安理工大学印刷包装与数字媒体学院
关键词
人体动作识别; 深度传感器; Kinect; 骨架关节点; 深度数据;
D O I
10.19322/j.cnki.issn.1006-4710.2015.03.001
中图分类号
TP391.41 [];
学科分类号
摘要
随着低成本深度传感器的发明,尤其是微软Kinect的出现,高分辨率的深度与视觉(RGB)感知数据被广泛使用,并为解决计算机视觉领域中的基本问题开拓了新的机遇。本文针对基于深度信息的人体动作识别研究,首先提出了一种基于特征和数据类型的分类框架,并对最近几年提出的相关方法进行了全面回顾。随后,对文献中描述的算法进行了性能对比分析,同时对所引用的公共测试数据集进行了总结。最后,笔者对未来的研究方向进行了讨论并给出了相关建议。
引用
收藏
页码:253 / 264+250 +250
页数:13
相关论文
共 8 条
[1]
Fusing Multiple Features for Depth-Based Action Recognition [J].
Zhu, Yu ;
Chen, Wenbin ;
Guo, Guodong .
ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY, 2015, 6 (02)
[2]
Accurate 3D action recognition using learning on the Grassmann manifold [J].
Slama, Rim ;
Wannous, Hazem ;
Daoudi, Mohamed ;
Srivastava, Anuj .
PATTERN RECOGNITION, 2015, 48 (02) :556-567
[3]
Evolutionary joint selection to improve human action recognition with RGB-D devices [J].
Andre Chaaraoui, Alexandros ;
Ramon Padilla-Lopez, Jose ;
Climent-Perez, Pau ;
Florez-Revuelta, Francisco .
EXPERT SYSTEMS WITH APPLICATIONS, 2014, 41 (03) :786-794
[4]
Effective 3D action recognition using EigenJoints [J].
Yang, Xiaodong ;
Tian, YingLi .
JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, 2014, 25 (01) :2-11
[5]
Exploring the Trade-off Between Accuracy and Observational Latency in Action Recognition [J].
Ellis, Chris ;
Masood, Syed Zain ;
Tappen, Marshall F. ;
LaViola, Joseph J., Jr. ;
Sukthankar, Rahul .
INTERNATIONAL JOURNAL OF COMPUTER VISION, 2013, 101 (03) :420-436
[6]
Motion recognition using local auto-correlation of space–time gradients.[J].Takumi Kobayashi;Nobuyuki Otsu.Pattern Recognition Letters.2012, 9
[7]
A survey of vision-based methods for action representation; segmentation and recognition.[J].Daniel Weinland;Remi Ronfard;Edmond Boyer.Computer Vision and Image Understanding.2010, 2
[8]
Efficient content-based retrieval of motion capture data [J].
Müller, M ;
Röder, T ;
Clausen, M .
ACM TRANSACTIONS ON GRAPHICS, 2005, 24 (03) :677-685