[1]王淑青,刘逸凡.基于改进自顶向下的行人运动预测方法[J].湖北工业大学学报,2020,(4):6-10.
　WANG Shuqing,LIU Yifan.A Pedestrian Motion Prediction Method Based on Improved Two-step Framework[J].,2020,(4):6-10.
点击复制

基于改进自顶向下的行人运动预测方法()
分享到：

《湖北工业大学学报》[ISSN:1003-4684/CN:42-1752/Z]

卷:
期数:
2020年第4期
页码:
6-10
栏目:
湖北工业大学学报
出版日期:
2020-08-31

文章信息/Info

Title:
A Pedestrian Motion Prediction Method Based on Improved Two-step Framework
文章编号:
1003-4684(2020)04-0006-05
作者:
王淑青;  刘逸凡
湖北工业大学电气与电子工程学院， 湖北 武汉 430068
Author(s):
WANG Shuqing; LIU Yifan
School of Electrical and Electronic Engin., Hubei Univ. of Tech., Wuhan 430068,China
关键词:
自顶向下法;  空间变换网络;  单人姿势检测;  光流处理
Keywords:
two-step framework;  Spatial transformation network;  single person posture detection;  optical flow
分类号:
TP242.2
文献标志码:
A
摘要:
针对传统的在复杂环境中多人姿态检测的自顶向下算法存在依赖人体框检测而导致的单人姿态检测错误的现象，提出一种基于改进的自顶向下的多人姿态检测并应用于行人运动预测的方法。该算法通过处理空间变换网络、单人姿势检测、反空间变换网络处理后的图像提取人体的骨点和姿势，经光流处理和长短期记忆神经网络的训练，预测行人接下来的动作。所述算法与6种经典多人姿态检测算法对比分析，实验结果表明：该方法得到的多人姿态检测图像准确，无冗杂人体框，无冗杂骨点，无骨骼交叉情况，行人运动预测效果良好。
Abstract:
Aiming at the problem that traditional two-step framework algorithm of multi-person posture detection in complex environment relies on human frame detection, this paper proposes an improved two-step framework algorithm of multi-person posture detection method and applies it to pedestrian motion prediction. The algorithm extracts human bones and postures from images processed by spatial transformation network, single-person posture detection and anti-spatial transformation network, and predicts pedestrian’s next actions through optical flow processing and training of long short-term memory network. Compared with six classical multi-person posture detection algorithms, the experimental results show that the multi-person posture detection image obtained by this method is accurate, without redundant human frame, no redundant bone points, no skeleton crossover, and the pedestrian motion prediction effect behaves well.

参考文献/References:

[1] Wei S E, Ramakrishna V,Kanade T,et al. Convolutional pose machines[C]∥Proceedings of the IEEE conference on Computer Vision and Pattern Recognition,2016: 4724-4732. 
[2] Fan X,Zheng K,Lin Y,et al. Combining local appearance and holistic view: Dual-source deep neural networks for human pose estimation[C]∥Proceedings of the IEEE conference on computer vision and pattern recognition,2015: 1347-1355.
[3] Sun M,Kohli P,Shotton J. Conditional regression forests for human pose estimation[C]∥2012 IEEE Conference on Computer Vision and Pattern Recognition. IEEE,2012: 3394-3401.
[4] Ladicky L,Torr P H S,Zisserman A. Human pose estimation using a joint pixel-wise and part-wise formulation[C]∥proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2013: 3578-3585.
[5] Gkioxari G,Hariharan 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.
[6] Insafutdinov E,Andriluka M,Pishchulin L,et al. Arttrack: Articulated multi-person tracking in the wild[C]∥Proceedings of the IEEE conference on computer vision and pattern recognition,2017: 6457-6465.
[7] Chen X,Yuille A L. Parsing occluded people by flexible compositions[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2015: 3945-3954.
[8] Insafutdinov E,Pishchulin L,Andres B,et al. Deepercut: A deeper,stronger,and faster multi-person pose estimation model[C]∥European Conference on Computer Vision. Springer,Cham,2016: 34-50. 
[9] Toshev A,Szegedy C. Deeppose: Human pose estimation via deep neural networks[C]∥Proceedings of the IEEE conference on computer vision and pattern recognition,2014: 1653-1660.
[10] Yang W,Li S,Ouyang W,et al. Learning feature pyramids for human pose estimation[C]∥proceedings of the IEEE international conference on computer vision,2017: 1281-1290.
[11] Pishchulin L,Jain A,Andriluka M,et al. Articulated people detection and pose estimation: Reshaping the future[C]∥2012 IEEE Conference on Computer Vision and Pattern Recognition. IEEE,2012: 3178-3185. 
[12] Dong G,Li J. Efficient mining of emerging patterns: Discovering trends and differences[C]∥Proceedings of the fifth ACM SIGKDD international conference on Knowledge discovery and data mining. 1999: 43-52.
[13] Pishchulin L,Insafutdinov E,Tang S,et al. Deepcut: Joint subset partition and labeling for multi person pose estimation[C]∥Proceedings of the IEEE conference on computer vision and pattern recognition,2016: 4929-4937. 
[14] Zhang S,Ding Y,Hao K,et al. An efficient two-step solution for vision-based pose determination of a parallel manipulator[J]. Robotics and Computer-Integrated Manufacturing,2012,28(2): 182-189. 
[15] Liu W,Anguelov D,Erhan D,et al. Ssd: Single shot multibox detector[C]∥European conference on computer vision. Springer,Cham,2016: 21-37. 
[16] Wei S E,Ramakrishna V,Kanade T,et al. Convolutional pose machines[C]∥Proceedings of the IEEE conference on Computer Vision and Pattern Recognition,2016: 4724-4732. 
[17] Horn BKP,Brian G. Schunck. Determining optical flow[J]. Artificial Intelligence,1980,17(1-3):185-203.
[18] Greff K,Srivastava R K,Koutník J,et al. LSTM: A search space odyssey[J]. IEEE transactions on neural networks and learning systems,2016,28(10): 2222-2232. 
[19] Iqbal U,Gall J. Multi-person pose estimation with local joint-to-person associations[C]∥European Conference on Computer Vision. Springer,Cham,2016: 627-642. 
[20] Levinkov E,Uhrig J,Tang S,et al. Joint graph decomposition & node labeling: Problem,algorithms,applications[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2017: 6012-6020. 
[21] Cao Z,Simon T,Wei S E,et al. Realtime multi-person 2d pose estimation using part affinity fields[C]∥Proceedings of the IEEE conference on computer vision and pattern recognition,2017: 7291-7299.
[22] Newell A,Huang Z,Deng J. Associative embedding: End-to-end learning for joint detection and grouping[C]∥Advances in neural information processing systems,2017: 2277-2287.
[23] Andriluka M,Pishchulin L,Gehler P,et al,2d human pose estimation: New benchmark and state of the art analysis[C]∥Proceedings of the IEEE Conference on computer Vision and Pattern Recognition,2014: 3686-3693.

备注/Memo

备注/Memo:
［收稿日期］ 2019-07-28
［基金项目］ 国家自然科学基金青年基金项目（61603127）
［第一作者］ 王淑青(1969-)， 女，河北衡水人，理学博士，湖北工业大学教授，研究方向为智能检测与控制，系统分析与集成
更新日期/Last Update: 2020-08-16