• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学 ›› 2021, Vol. 43 ›› Issue (01): 105-111.

• 图形与图像 • 上一篇    下一篇

基于可变形卷积神经网络的人体动作识别

王雪娇,智敏   

  1. (内蒙古师范大学计算机科学技术学院,内蒙古 呼和浩特 010020)

  • 收稿日期:2020-02-28 修回日期:2020-04-20 接受日期:2021-01-25 出版日期:2021-01-25 发布日期:2021-01-22
  • 基金资助:
    内蒙古自治区自然科学基金(2018MS06008);内蒙古师范大学研究生科研创新基金(内蒙古自治区研究设计研究生教育创新计划)(CXJJS19150)

Human motion recognition based on deformable convolutional neural network

WANG Xue-jiao,ZHI Min   

  1. (College of Computer Science and Technology,Inner Mongolia Normal University,Hohhot 010020,China)

  • Received:2020-02-28 Revised:2020-04-20 Accepted:2021-01-25 Online:2021-01-25 Published:2021-01-22

摘要: 针对复杂场景中人体动作识别准确率不高的问题,构建了一种基于可变形卷积网络(DCN)与可变形部件模型(DPM)融合改进的人体动作识别系统。首先将DPM的部件滤波器由5个增加到8个,并结合分支定界算法共同将准确率提高约11个百分点,速度提高3倍左右;其次利用DCN根据人体动作进行感兴趣点采样;然后将改进的DPM与DCN在可变形池化前进行融合;最后通过全连接层对输入数据进行动作的识别。实验结果表明,此系统能够在人体动作数据集上更快、更准确地得到识别结果。

关键词: 人体动作识别, 可变形卷积, 可变形感兴趣池化, 可变形部件模型算法, 卷积神经网络, 分支定界算法

Abstract: To solve the problem of low accuracy of human motion recognition in complex scenes, an improved human motion recognition system based on deformable convolution network (DCN) and deformable part model (DPM) is constructed. Firstly, the number of the DPM component filters are increased from 5 to 8, and the branch and bound method is combined to improve the accuracy by about 11% and the speed by about 3 times. Secondly, DCN is used to sample the points of interest according to the movements of human body. Then, the improved DPM and DCN are fused before deformable pool- ing. Finally, the input data is identified by the full connection layer.Experimental results show that this method can identify the results more quickly and accurately on the human movement dataset.


Key words: human motion recognition, deformable convolution, deformable interest pooling, deformable part model algorithm, convolutional neural network, branch and bound algorithm