帮助 关于我们

返回检索结果

基于DeepPose和Faster RCNN的多目标人体骨骼节点检测算法
Human body joint nodes detection based on DeepPose and Faster RCNN

查看参考文献15篇

余保玲 1   虞松坤 1   孙耀然 2   杨振 3   傅旭波 1 *  
文摘 近年来,随着计算机视觉技术的不断发展,深度学习技术在人体关节节点检测中得到了很好的应用。但是由于人体关节结构复杂,关节之间存在相互依赖的关系和互相遮挡的问题,因此人体骨骼节点检测依然是一个极具挑战的任务。传统的模型难以预测多个目标的骨骼节点,为了解决这个问题,提出一种基于Faster RCNN和DeepPose相结合的方法,首先通过Faster RCNN检测出包含人体的感兴趣区域,并将该区域作为改进的DeepPose算法的输入,使其能够处理多目标关节节点检测的问题。实验表明,该算法在MPII数据集的手腕、膝盖两种关键节点检测上均取得最好结果,比之前的最好结果各提升1.2%和0.3%,在全部的关键节点检测上PCKh为87.6%。
其他语种文摘 Human body joint nodes detection is a considerably challenging task which has drawn enormous attention in the field of computer vision recently.The challenges of this task include:coping with the complex structure of human body joints,denoting the interdependence between joint nodes,and dealing with the sheltered and overlapped body joint nodes.Among the common solutions to this task,the models based on deep learning are widely applied and provide useful results.However,the existing models have following drawbacks:1)comparatively low accuracy in prediction;2)poor performance in multi-objective tasks.In our work,we proposed a novel method aiming at more satisfactory results.We firstly detect the relevant regions of human body with Faster RCNN,and then input the regions into a modified DeepPose algorithm.We achieve the state-of-theart results in the detection of the wrist and knee on MPII dataset,improving 1.2% and 0.3% in PCKh,respectively.The total PCKh is 87.6% on MPII dataset.
来源 中国科学院大学学报（中英文） ,2020,37(6):828-834 【核心库】
DOI 10.7523/j.issn.2095-6134.2020.06.015
关键词 Faster RCNN ; DeepPose ; 人体关节节点检测
地址

1. 浙江大学公共体育与艺术部, 杭州, 310058  

2. 浙江大学光电科学与工程学院, 杭州, 310058  

3. 中国科学院自动化研究所, 北京, 100190

语种 中文
文献类型 研究性论文
ISSN 2095-6134
学科 自动化技术、计算机技术
基金 国家重点研发计划项目 ;  胶州人工智能产业技术研究院开放课题资助
文献收藏号 CSCD:6849368

参考文献 共 15 共1页

1.  Tompson J J. Joint training of a convolutional network and a graphical model for human pose estimation. Advances in Neural Information Processing Systems,2014:1799-1807 CSCD被引 13 次    
2.  Toshev A. Human pose estimation via deep neural networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2014:24-27 CSCD被引 1 次    
3.  Sapp B. Modec: multimodal decomposable models for human pose estimation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2013:3674-3681 CSCD被引 5 次    
4.  Gkioxari G. Chained predictions using convolutional neural networks. European Conference on Computer Vision,2016:728-743 CSCD被引 1 次    
5.  He K. Spatial pyramid pooling in deep convolutional networks for visual recognition. IEEE Transactions On Pattern Analysis and Machine Intelligence,2015,37(9):1904-1916 CSCD被引 1346 次    
6.  Girshick R. Fast r-cnn. Proceedings of the IEEE International Conference on Computer Vision,2015:1440-1448 CSCD被引 717 次    
7.  Gkioxari G. Contextual action recognition with r * cnn. Proceedings of the IEEE International Conference on Computer Vision,2015:1080-1088 CSCD被引 7 次    
8.  He K. Mask r-cnn. Proceedings of the IEEE International Conference on Computer Vision,2017:2961-2969 CSCD被引 338 次    
9.  Ren S. Faster r-cnn: Towards realtime object detection with region proposal networks. Advances in Neural Information Processing Systems,2015:91-99 CSCD被引 4396 次    
10.  Everingham M. The pascal visual object classes (voc) challenge. International Journal of Computer Vision,2010,88(2):303-338 CSCD被引 753 次    
11.  Johnson S. Clustered pose and nonlinear appearance models for human pose estimation. BMVC,2010 CSCD被引 2 次    
12.  . MPI human pose dafaset,2019 CSCD被引 1 次    
13.  Tompson J. Efficient object localization using convolutional networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2015:648-656 CSCD被引 15 次    
14.  Pishchulin L. Deepcut: joint subset partition and labeling for multi person pose estimation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2016:4929-4937 CSCD被引 14 次    
15.  Wei S E. Convolutional pose machines. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2016:4724-4732 CSCD被引 37 次    
引证文献 3 篇

1 李平 基于两种分类标准的目标检测算法综述 计算机应用研究,2021,38(9):2582-2589
CSCD被引 2 次

2 盛晓光 WILS:面向学业预警的非均衡增量式学习方法 中国科学院大学学报（中英文）,2023,40(3):422-432
CSCD被引 0 次

显示所有3篇文献

论文科学数据集
PlumX Metrics
相关文献

 作者相关
 关键词相关
 参考文献相关

版权所有 ©2008 中国科学院文献情报中心 制作维护：中国科学院文献情报中心
地址：北京中关村北四环西路33号 邮政编码：100190 联系电话：(010)82627496 E-mail:cscd@mail.las.ac.cn 京ICP备05002861号-4 | 京公网安备11010802043238号