Latent variable pictorial structure for human pose estimation on depth images

NEUROCOMPUTING

卷:203

Article

He, Li¹  Wang, Guijin¹  Liao, Qingmin²  Xue, Jing-Hao³ 

[1] Tsinghua Univ, Dept Elect Engn, Beijing 100084, Peoples R China

[2] Grad Sch Shenzhen, Dept Elect Engn, Tsinghua Campus, Shenzhen 518055, Peoples R China

[3] UCL, Dept Stat Sci, London WC1E 6BT, England

关键词: Pose estimation;    Pictorial structure;    Latent variable;    Body silhouette;    Regression forest;    Depth images;   

DOI  :  10.1016/j.neucom.2016.04.009

来源: Elsevier

【 摘 要 】

Prior models of human pose play a key role in state-of-the-art techniques for monocular pose estimation. However, a simple Gaussian model cannot represent well the prior knowledge of the pose diversity on depth images. In this paper, we develop a latent variable-based prior model by introducing a latent variable into the general pictorial structure. Two key characteristics of our model (we call Latent Variable Pictorial Structure) are as follows: (1) it adaptively adopts prior pose models based on the estimated value of the latent variable; and (2) it enables the learning of a more accurate part classifier. Experimental results demonstrate that the proposed method outperforms other state-of-the-art methods in recognition rate on the public datasets. (C) 2016 Published by Elsevier B.V.

【 授权许可】

Free