Human pose estimation based on configuration constraints and trajectory optimization

Qing-wu LI; Shu-ya XI; Tian WANG; Yun-peng MA; Liang-ji ZHOU

doi:10.3788/OPE.20172504.1060

您当前的位置：

首页 >

文章列表页 >

Human pose estimation based on configuration constraints and trajectory optimization

Information Sciences | 更新时间：2020-07-07

- Human pose estimation based on configuration constraints and trajectory optimization
- Optics and Precision Engineering Vol. 25, Issue 4, Pages: 1060-1069(2017)
- 作者机构：
  
  1.河海大学物联网工程学院, 江苏常州 213022
  2.常州市传感网与环境感知重点实验室, 江苏常州 213022
- 作者简介：
- 基金信息：
- DOI：10.3788/OPE.20172504.1060
  CLC： TP391
- Received：24 November 2016，
  
  Accepted：16 January 2017，
  
  Published：25 April 2017
- 稿件说明：
移动端阅览
Qing-wu LI, Shu-ya XI, Tian WANG, et al. Human pose estimation based on configuration constraints and trajectory optimization[J]. Optics and precision engineering, 2017, 25(4): 1060-1069.
DOI：

Qing-wu LI, Shu-ya XI, Tian WANG, et al. Human pose estimation based on configuration constraints and trajectory optimization[J]. Optics and precision engineering, 2017, 25(4): 1060-1069. DOI： 10.3788/OPE.20172504.1060.

摘要

基于混合部件模型的人体姿态估计方法忽视了人体结构的对称位姿约束关系，从而导致对称部件容易被重复检测、人体姿态估计准确率较低，为此，提出一种基于位姿约束与轨迹寻优的姿态估计新方法。首先估计人体单部件和对称部件在单帧图像中的多个合理位置，利用对称部件之间的位姿约束关系构建标识部件。然后根据单部件和标识部件各自的目标优化函数，通过动态规划算法反复迭代获得初始轨迹候选集，再结合轨迹的全局特征剔除检测得分较低的运动轨迹。最后引入树形合约模型，联系时空上下文信息，准确求解出视频序列光滑且兼容的最优轨迹。在N-best、Outdoor Pose和Scene数据集中的实验结果表明，对于存在背景复杂、运动模糊、部件遮挡等问题的视频序列中，该方法平均姿态估计准确率达87%以上，有效减少了对称部件的误判，提高了视频中人体姿态估计的准确率。

Abstract

Because of ignoring the configuration constraints between symmetric body parts

the human pose estimation methods based on mixtures of parts may lead to a repetitive detection of symmetrical body parts and a low pose estimation accuracy. Therefore

a kind of new pose estimation method on the basis of pose constraint and trajectory optimization was put forward. Firstly

numerous reasonable locations of single part and symmetric parts of human in single-frame image should be estimated

and identification part should be constructed by utilizing pose constraint relationship among symmetric parts. Then initial trajectory candidates set shall be gained through repeated iteration of dynamic programming algorithm according to respective target optimization function of single part and identification part. Movement trajectory with relatively low detection score was removed by combining with global feature of trajectory. Finally

smooth and compatible optimal trajectory of video sequence was correctly solved by introducing tree-based contract model and combining with contextual spatio-temporal information. Experimental result in N-best

Outdoor Pose and Scene dataset shows that in video sequence with complex background

blur movement and part blocking problems

average pose estimation accuracy of proposed method is greater than 87%

which reduces erroneous judgment of symmetric parts effectively and improves human pose estimation accuracy in video.

关键词

Keywords

references

李静宇, 刘艳滢, 田睿, 等.视频监控系统中的概率模型单目标跟踪框架[J].光学精密工程, 2015, 23(7):2093-2099.

LI J Y, LIU Y Y, TIAN R, et al.. Probabilistic model single target tracking framework for video surveillance system[J]. Opt. Precision Eng., 2015, 23(7):2093-2099.(in Chinese)

李玉峰, 李广泽, 谷绍湖, 等.基于区域分块与尺度不变特征变换的图像拼接算法[J].光学精密工程, 2016, 24(5):1197-1205.

LI Y F, LI G Z, GU SH H, et al.. Image mosaic algorithm based on area blocking and SIFT[J]. Opt. Precision Eng., 2016, 24(5):1197-1205.(in Chinese)

胡梦婕, 魏振忠, 张广军.基于对象性测度估计和霍夫森林的目标检测方法[J].红外与激光工程, 2015, 44(6):1936-1941.

HU M J, WEI ZH ZH, ZHANG G J. Object detection method based on objectness estimation and Hough forest[J]. Infrared and Laser Engineering, 2015, 44(6):1936-1941.(in Chinese)

杨磊, 任龙, 刘庆, 等.基于FPGA的大视场图像实时拼接技术的研究与实现[J].红外与激光工程, 2015, 44(6):1929-1935.

YANG L, REN L, LIU Q, et al.. Research and implementation of large field image real-time mosaic technology based on FPGA[J]. Infrared and Laser Engineering, 2015, 44(6):1929-1935.(in Chinese)

田国会, 尹建芹, 韩旭, 等.一种基于关节点信息的人体行为识别新方法[J].机器人, 2014, 36(3):285-292.

TIAN G H, YIN J Q, HAN X, et al.. A novel human activity recognition method using joint points information[J]. Robot, 2014, 36(3):285-292.(in Chinese)

韩贵金, 朱虹.基于HOG和颜色特征融合的人体姿态估计[J].模式识别与人工智能, 2014, 27(9):769-777.

HAN G J, ZHU H. Human pose estimation based on fusion of HOG and color feature[J]. PR & AI., 2014, 27(9):769-777.(in Chinese)

EICHNER M, MARIN-JIMENEZ M, ZISSERMAN A, et al.. 2D articulated human pose estimation and retrieval in (almost) unconstrained still images[J]. International Journal of Computer Vision, 2012, 99(2):190-214.

YANG Y, RAMANAN D. Articulated human detection with flexible mixtures of parts[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(12):2878-2890.

GKIOXARI G, HARIHRAN B, GIRSHICK R, et al.. Using k-poselets for detecting people and localizing their keypoints[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2014:3582-3589.

LÓPEZ-QUINTERO M I, MARÍN-JIMÉNEZ M J, MUÑOZ-SALINAS R, et al.. Stereo pictorial structure for 2D articulated human pose estimation[J]. Machine Vision and Applications, 2016, 27(2):157-174.

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.

OUYANG W, CHU X, WANG X. Multi-source deep learning for human pose estimation[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2014:2337-2344.

BUEHLER P, EVERINGHAM M, HUTTENLOCHER D P, et al.. Upper body detection and tracking in extended signing sequences[J]. International Journal of Computer Vision, 2011, 95(2):180-197.

PARK D, RAMANAN D. N-best maximal decoders for part models[C]. Proceedings of the IEEE International Conference on Computer Vision , 2011:2627-2634.

马淼, 李贻斌.基于多级动态模型的2维人体姿态估计[J].机器人, 2016, 38(5):578-587.

MA M, LI Y B. 2D human pose estimation using multi-level dynamic model[J]. Robot, 2016, 38(5):578-587.(in Chinese)

RAMAKRISHNA V, KANADE T, SHEIKH Y. Tracking human pose by tracking symmetric parts[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2013:3728-3735.

RAMAKRISHNA V, MUNOZ D, HEBERT M, et al.. Pose machines:articulated pose estimation via inference machines[C]. European Conference on Computer Vision , 2014:33-47.

TOKOLA R, CHOI W, SAVARESE S. Breaking the chain:liberation from the temporal Markov assumption for tracking human poses[C]. Proceedings of the IEEE International Conference on Computer Vision , 2013:2424-2431.

BATRA D, YADOLLAHPOUR P, GUZMAN-RIVERA A, et al.. Diverse M-best solutions in markov random fields[C]. European Conference on Computer Vision, Springer Berlin Heidelberg , 2012:1-16.

SAPP B, WEISS D, TASKAR B. Parsing human motion with stretchable models[C]. Computer Vision and Pattern Recognition, IEEE , 2011:1281-1288.

CHERIAN A, MAIRAL J, ALAHARI K, et al.. Mixing body-part sequences for human pose estimation[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition . 2014:2361-2368.

FERRARI V, MARIN-JIMENEZ M, ZISSERMAN A. Progressive search space reduction for human pose estimation[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2008:1-8.

Views

599

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Novel approach to generating optimal smooth trajectory for a manipulator

Related Author

CHEN Zhong-ze

YAN Guo-zheng

LIN Liang-ming

CAI Hong

Related Institution

. 820 Lab, College of Electronics and Information, Shanghai Jiaotong University, Shanghai

AI问答

Address：No.3888 Dong Nanhu Road, Changchun, Jilin, China Postal code：130033
Tel：0431-86176855 Email：gxjmgc@ciomp.ac.cn
Technical support is provided by Beijing Founder electronics co., LTD 吉ICP备11002662号-17 京公网安备11010802024621
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰