|本期目录/Table of Contents|

[1]肖澳文,刘 军*,张苏沛,等.基于CNN的三维人体姿态估计方法[J].武汉工程大学学报,2019,(02):168-172.[doi:10. 3969/j. issn. 1674?2869. 2019. 02. 013]
 XIAO Aowen,LIU Jun*,ZHANG Supei,et al.Three-Dimensional Human Pose Estimation Based on Convolution Neural Network[J].Journal of Wuhan Institute of Technology,2019,(02):168-172.[doi:10. 3969/j. issn. 1674?2869. 2019. 02. 013]
点击复制

基于CNN的三维人体姿态估计方法(/HTML)
分享到:

《武汉工程大学学报》[ISSN:1674-2869/CN:42-1779/TQ]

卷:
期数:
2019年02期
页码:
168-172
栏目:
机电与信息工程
出版日期:
2019-04-18

文章信息/Info

Title:
Three-Dimensional Human Pose Estimation Based on Convolution Neural Network
文章编号:
20190213
作者:
肖澳文12刘 军*12张苏沛12杜 壮12孙思琪12
1. 智能机器人湖北省重点实验室(武汉工程大学),湖北 武汉 430205;2. 武汉工程大学计算机科学与工程学院,湖北 武汉 430205
Author(s):
XIAO Aowen12 LIU Jun*12 ZHANG Supei12 DU Zhuang12 SUN Siqi12
1. Hubei Key Laboratory of Intelligent Robot(Wuhan Institute of Technology), Wuhan 430205, China;2. School of Computer Science & Engineering, Wuhan Institute of Technology, Wuhan 430205, China
关键词:
三维人体姿态估计卷积神经网络关节点
Keywords:
three-dimensional human pose estimation convolution neural network joint points
分类号:
TP317.4
DOI:
10. 3969/j. issn. 1674?2869. 2019. 02. 013
文献标志码:
A
摘要:
针对传统三维人体姿态估计受遮挡限制的问题,提出一种基于卷积神经网络(CNN)的三维人体姿态估计方法。首先,实验模型系统采用了几段单目视频为输入源进行人体姿态识别。相对于传统的人体姿态估计方法,改进了一种顺序化的卷积神经网络用于提取人体空间信息和纹理信息。并通过对视频中人体的二维姿态估计,找出了人体头部和四肢关节点的精确位置。最后,通过投影关节点到三维空间,估计出每个人的三维姿态。实验结果表明,本文方法相比传统的姿态估计算法在人体行为上的测试平均误差从98.53 mm降低至92.88 mm,对于视频中的人体三维姿态估计有更优的精度。
Abstract:
To solve the problem that the traditional three-dimensional human pose estimation performance was limited by occlusion, this paper presents a three-dimensional human pose estimation method based on convolution neural network. Firstly, some monocular videos were used as the inputs to recognize the human body postures in the experiment model. Secondly, a sequential convolution neural network was constructed to extract the spatial and texture information of human body. Thirdly, the exact position of the joint points of the head and body was found through two-dimensional human pose estimation in the video. Finally, the three-dimensional pose of each person was estimated by projecting the correlation node to the three-dimensional space. The experimental results show that the mean error reduces from 98.53 mm to 92.88 mm compared with the traditional human pose estimation algorithm, and our method has higher precision in the three-dimensional human pose estimation in the testing video.

参考文献/References:

[1] 辛义忠,邢志飞. 基于Kinect的人体动作识别方法[J]. 计算机工程与设计,2016,37(4):1056-1061.[2] 赵海峰,费婷婷,王文中,等. 结合个性化建模和深度数据的三维人体姿态估计[J]. 计算机系统应用,2016,25(11):118-125.[3] 陈起凤,刘军,李威,等. 三维重建中线段匹配方法的研究[J]. 武汉工程大学学报,2018,40(4):446-450.[4] 郭盛威,章秀华,范艳,等. 三维重建表面几何特征的提取与参数测量计算[J]. 武汉工程大学学报, 2016, 38(2):185-188.[5] 刘军,李娜,刘鹏. 双目视觉立体标定方法的改进[J]. 武汉工程大学学报,2013,35(10):68-73.[6] 周志华,陈世福. 神经网络集成[J]. 计算机学报,2002,25(1):1-8.[7] 张苏沛,刘军,肖澳文,等. 基于卷积神经网络的验证码识别[J]. 武汉工程大学学报,2019,41(1):89-92.[8] 王钰清,陆文凯,刘金林,等. 基于数据增广和CNN的地震随机噪声压制[J]. 地球物理学报,2019,62(1):421-433.[9] 吴和保,李晓微,龙玉阳,等. 人工神经网络快速预测蠕墨铸铁的性能[J]. 武汉工程大学学报,2013,35(10):63-67.[10] 李天峰. 基于多媒体技术的三维人物图像动态重构[J]. 现代电子技术,2018,41(9):68-71.[11] BOGO F , KANAZA W A , LASSNER C , et al. Keep it SMPL: automatic estimation of 3D human pose and shape from a single image [J]. Springer International Publishing,2016,10(6): 561-578.[12] ZHOU X , ZHU M , LEONARDOS S , et al. Sparseness meets deepness: 3D human pose estimation from monocular video [J]. IEEE Conference on Computer Vision and Pattern Recognition,2016,537(1): 4966-4975.[13] PAVLAKOS G , ZHOU X , DERPANIS K G , et al. Coarse-to-fine volumetric prediction for single-image 3D human pose [J]. IEEE Conference on Computer Vision and Pattern Recognition, 2016,139(1):1263- 1272.[14] WEI S E , RAMAKRISHNA V , KANADE T , et al. Convolutional pose machines [J]. IEEE Conference on Computer Vision and Pattern Recognition, 2016,511(1): 4724-4732.[15] IONESCU C , PAPAVA D , OLARU V , et al. Human3.6M: large scale datasets and predictive methods for 3D human sensing in natural environments [J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2014, 36(7):1325-1339.[16] 杨博雄,杨雨绮. 利用PCA进行深度学习图像特征提取后的降维研究[J]. 计算机系统应用,2019,28(1):279-283.微加热器上不同工质的气泡动力学实验

相似文献/References:

[1]汪家明,卢 涛*.多尺度残差深度神经网络的卫星图像超分辨率算法[J].武汉工程大学学报,2018,40(04):440.[doi:10. 3969/j. issn. 1674?2869. 2018. 04. 018]
 WANG Jiaming,LU Tao *. Satellite Imagery Super-Resolution Algorithm via Multi-Scale Residual Deep Neural Network[J].Journal of Wuhan Institute of Technology,2018,40(02):440.[doi:10. 3969/j. issn. 1674?2869. 2018. 04. 018]
[2]张苏沛,刘 军*,肖澳文,等.基于卷积神经网络的验证码识别[J].武汉工程大学学报,2019,(01):89.[doi:10. 3969/j. issn. 1674?2869. 2019. 01. 015]
 ZHANG Supei,LIU Jun*,XIAO Aowen,et al.CAPTCHA Recognition Based on Convolutional Neural Network[J].Journal of Wuhan Institute of Technology,2019,(02):89.[doi:10. 3969/j. issn. 1674?2869. 2019. 01. 015]
[3]陈希彤,卢 涛*.基于全局深度分离卷积残差网络的高效人脸识别算法[J].武汉工程大学学报,2019,(03):276.[doi:10. 3969/j. issn. 1674-2869. 2019. 03. 014]
 CHEN Xitong,LU Tao *.Efficient Face Recognition Algorithm Using Global Deep Separable Convolutional and Residual Network[J].Journal of Wuhan Institute of Technology,2019,(02):276.[doi:10. 3969/j. issn. 1674-2869. 2019. 03. 014]
[4]王丽亚,刘昌辉*,蔡敦波,等.基于CNN-BiLSTM网络引入注意力模型的文本情感分析[J].武汉工程大学学报,2019,(04):386.[doi:10. 3969/j. issn. 1674?2869. 2019. 04. 016]
 WANG Liya,LIU Changhui*,CAI Dunbo,et al.Text Sentiment Analysis Based on CNN-BiLSTM Network and Attention Model[J].Journal of Wuhan Institute of Technology,2019,(02):386.[doi:10. 3969/j. issn. 1674?2869. 2019. 04. 016]
[5]熊寒颖,鲁统伟*,闵 峰,等.基于单一神经网络的实时人脸检测[J].武汉工程大学学报,2019,(05):489.[doi:10. 3969/j. issn. 1674?2869. 2019. 05. 015]
 XIONG Hanying,LU Tongwei*,MIN Feng,et al.Real-Time Face Detection Based on Single Neural Network[J].Journal of Wuhan Institute of Technology,2019,(02):489.[doi:10. 3969/j. issn. 1674?2869. 2019. 05. 015]
[6]杜梦星,王彦伟*.基于CNN的突发事件预警系统的设计与实现[J].武汉工程大学学报,2020,42(02):207.[doi:10.19843/j.cnki.CN42-1779/TQ.201910016]
 DU Mengxing,WANG Yanwei*.Design and Implementation of Emergency Warning System Based on Convolution Neural Network[J].Journal of Wuhan Institute of Technology,2020,42(02):207.[doi:10.19843/j.cnki.CN42-1779/TQ.201910016]
[7]江满星,赵彤洲*,吴泽俊.基于目标形状卷积神经网络在舰船分类中的应用[J].武汉工程大学学报,2020,42(02):213.[doi:10.19843/j.cnki.CN42-1779/TQ.201911022]
 JIANG Manxing,ZHAO Tongzhou*,WU Zejun.Application of Convolution Neural Network Based on Target Shape in Ships and Warships Classification[J].Journal of Wuhan Institute of Technology,2020,42(02):213.[doi:10.19843/j.cnki.CN42-1779/TQ.201911022]

备注/Memo

备注/Memo:
收稿日期:2018-10-23基金项目:国家自然科学基金(61172150, 61803286);智能机器人湖北省重点实验室开放基金(HBIR 201802);武汉工程大学第十届研究生教育创新基金(CX2018197, CX2018200, CX2018212)作者简介:肖澳文,硕士研究生。E-mail:xiaoaowen@wit.edu.cn*通讯作者:刘 军,博士,副教授。E-mail:liujun@wit.edu.cn引文格式:肖澳文,刘军,张苏沛,等. 基于CNN的三维人体姿态估计方法[J]. 武汉工程大学学报,2019,41(2):168-172.
更新日期/Last Update: 2019-04-20