基于关键帧的人体动作识别方法

系统仿真学报 ›› 2015, Vol. 27 ›› Issue (10): 2401-2408.

基于关键帧的人体动作识别方法

石祥滨^1,2,3, 刘拴朋¹, 张德园¹

1.沈阳航空航天大学计算机学院,沈阳 110136;
2.沈阳航空航天大学辽宁通用航空重点实验室,沈阳 110136;
3.辽宁大学信息学院,沈阳 110036

收稿日期:2015-06-14 修回日期:2015-07-30 出版日期:2015-10-08 发布日期:2020-08-07
作者简介:石祥滨(1963-),男,辽宁,博士,教授,研究方向为虚拟现实,图像处理,网络游戏。
基金资助:
国家自然科学基金(61170185); 航空科学基金(2013ZC54011); 辽宁省博士启动基金 (20121034); 辽宁省教育厅资助项目(L2014070)

Human Action Recognition Method Based on Key Frames

Shi Xiangbin^1,2,3, Liu Shuanpeng¹, Zhang Deyuan¹

1. Department of Computer, Shenyang Aerospace University, Shenyang 110136, China;
2. Liaoning General Aviation Key Laboratory, Shenyang Aerospace University, Shenyang 110136, China;
3. College of Information, Liaoning University, Shenyang 110036, China

Received:2015-06-14 Revised:2015-07-30 Online:2015-10-08 Published:2020-08-07

摘要/Abstract

摘要： Kinect问世以来,越来越多的研究者开始研究基于深度信息和骨架信息的人体动作识别。为了提高动作识别的准确率和实时性,并且降低计算过程中的计算复杂度,提出了一个基于关键帧的骨架特征的人体动作识别方法。采用K-均值聚类算法对人体动作视频序列做聚类,通过聚类出的数据提取人体动作视频序列中的关键帧。提取关键帧中的关节点位置和人体刚体部分之间的骨架角度两种特征,利用SVM分类器对动作序列进行分类。在MSR-DailyActivity3D数据集上的实验结果表明,该方法具有较高的识别率,并且提高了实时性。

关键词: Kinect, 人体动作识别, 关键帧, k-means

Abstract: More and more researchers have begun to study the human action recognition based on depth information and skeleton information since the Kinect has been released. A method of human action recognition based on the skeleton feature of key frames is proposed in order to improve the accuracy and timeliness of the human action recognition, and reduce the computational complexity. The clustered data was obtained by using K-means clustering algorithm, and then the key frames were extracted by using the clustered data. Two features for human action recognition were extracted, one is the feature of the position of human joint, another is the feature of the skeleton angle between rigid body and rigid body. The sequence of action video was classified through the SVM classifier. This method leads to a more accurate recognition rate, and the real-time capability has been improved at the same time according to the result showed on the data set MSR-DailyActivity3D.

Key words: Kinect, human action recognition, key frames, k-means

中图分类号:

TP391.4

石祥滨, 刘拴朋, 张德园. 基于关键帧的人体动作识别方法[J]. 系统仿真学报, 2015, 27(10): 2401-2408.

Shi Xiangbin, Liu Shuanpeng, Zhang Deyuan. Human Action Recognition Method Based on Key Frames[J]. Journal of System Simulation, 2015, 27(10): 2401-2408.

参考文献

[1] 胡琼, 秦磊, 黄庆明. 基于视觉的人体动作识别综述[J]. 计算机学报, 2013, 36(12): 2512-2524.
[2] 徐光祐, 曹媛媛. 动作识别与行为理解综述[J]. 中国图象图形学报, 2009, 14(2): 189-195.
[3] J K Aggarwal, M S Ryoo. Human Activity Analysis: A Review [J]. ACM Computing Surveys(CSUR), 2011, 43(3): 16:1-16:43.
[4] Shotton J, Sharp T, Kipman A, et al.Real-time human pose recognition in parts from single depth images[J]. Communications of the ACM, 2013, 56(1): 116-124.
[5] Vemulapalli R, Arrate F, Chellappa R.Human action recognition by representing 3d skeletons as points in a lie group[C]// Computer Vision and Pattern Recognition (CVPR), 2014 IEEE Conference on. USA: IEEE, 2014: 588-595.
[6] Meinard Müller.Information Retrieval for Music and Motion[M]. USA: Springer-Verlag New York, Inc., 2007.
[7] Wang J, Liu Z, Wu Y, et al.Mining actionlet ensemble for action recognition with depth cameras[C]//Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on. IEEE, 2012: 1290-1297.
[8] Hussein M E, Torki M, Gowayyed M A, et al.Human action recognition using a temporal hierarchy of covariance descriptors on 3d joint locations[C] //Proceedings of the Twenty-Third international joint conference on Artificial Intelligence. AAAI Press, 2013: 2466-2472.
[9] Thi-Lan Le, Minh-Quoc Nguyen.Human posture recognition using human skeleton provided by Kinect[C]// Computing, Management and Telecommunications (ComManTel), 2013 International Conference on. USA: IEEE, 2013: 340-345.
[10] Liu T, Song Y, Gu Y, et al. Human Action Recognition Based on Depth Images from Microsoft Kinect[C]// Intelligent Systems (GCIS), 2013 Fourth Global Congress on. USA: IEEE, 2013: 200-204.
[11] Müller M, Röder T.Motion templates for automatic classification and retrieval of motion capture data[C]//Proceedings of the 2006 ACM SIGGRAPH/Eurographics symposium on Computer animation. Eurographics Association, 2006: 137-146.
[12] Chih-Chung C, L Chih-Jen. LIBSVM: A library for support vector machines[J]. ACM Transactions on Intelligent Systems and Technology (TIST), 2011, 2(3): 1-27.
[13] Le T L, Nguyen M Q, Nguyen T T M. Human posture recognition using human skeleton provided by Kinect[C]// Computing, Management and Telecommunications (ComManTel), 2013 International Conference on. USA: IEEE, 2013: 340-345.

[1]	王沐晴, 张磊, 范秀敏, 骆晓萌, 朱文敏. VR外设驱动的虚拟人姿态优化仿真方法[J]. 系统仿真学报, 2022, 34(6): 1296-1303.
[2]	唐传谦, 刘志强, 苏毅俊, 刘晓静. 基于Kinect的皮影动作实时匹配研究[J]. 系统仿真学报, 2022, 34(3): 633-639.
[3]	赵瑛, 陆耀, 张健, 梁启弟, 龙炜. 基于深度神经网络的多视角人体动作识别[J]. 系统仿真学报, 2021, 33(5): 1019-1030.
[4]	李琪, 王向东, 李华. 基于双Kinect传感器的三维人体姿态跟踪方法[J]. 系统仿真学报, 2020, 32(8): 1446-1454.
[5]	王闯, 张勇, 李学贵, 董宏丽. 改进粒子群优化算法及其在聚类分析中应用[J]. 系统仿真学报, 2020, 32(8): 1577-1587.
[6]	吴敬兵, 唐汉卿, 胥军. 水泥窑协同处置生活垃圾的燃烧特性分析优化[J]. 系统仿真学报, 2020, 32(1): 35-43.
[7]	王杰, 王艳. 基于量子遗传聚类算法的质量控制方法[J]. 系统仿真学报, 2019, 31(12): 2591-2599.
[8]	徐连瑞, 张锦明. 基于Kinect的室内环境建模[J]. 系统仿真学报, 2019, 31(12): 2643-2651.
[9]	纪显俐, 丁宇龙, 杨双华. 智能沙盘地理教学演示系统[J]. 系统仿真学报, 2019, 31(12): 2816-2828.
[10]	李鸣晓, 庚琦川, 莫红, 吴威, 周忠. 基于片段关键帧的视频行为识别方法[J]. 系统仿真学报, 2018, 30(7): 2787-2793.
[11]	马楠, 石祥滨, 代钦, 刘翠微, 刘芳. 一种音乐舞蹈视频关键帧提取方法[J]. 系统仿真学报, 2018, 30(7): 2801-2807.
[12]	魏毅, 夏时洪. 物理逼真的人体空翻运动合成[J]. 系统仿真学报, 2018, 30(6): 2190-2197.
[13]	张新良, 李阳, 赵运基. 一种基于时空切片的SLAM关键帧提取方法[J]. 系统仿真学报, 2018, 30(5): 1724-1729.
[14]	戴月明, 王明慧, 张明, 王艳. SVD优化初始簇中心的K-means中文文本聚类算法[J]. 系统仿真学报, 2018, 30(10): 3835-3842.
[15]	何鹏, 胡少军, 何东健. 基于Kinect的树运动轨迹获取与振动频率分析[J]. 系统仿真学报, 2017, 29(7): 1411-1418.