Human Pose Estimation Based on Deep Neural Network

This paper aimed to single pose estimation of two-dimensional image based on the deep neural network. We constructed a Stacked Hourglass network structure on the training data set, regress the precise pixel location of human body joint point through heat map with different activation function is analyzed and compared, which had certain reference value to the improvement on the final performance. At the same time, this paper analyzes the theory and application of multi-scale and analyzes the structure and advantages of network.

[1]  Peiyun Hu,et al.  Bottom-Up and Top-Down Reasoning with Hierarchical Rectified Gaussians , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Peter V. Gehler,et al.  Strong Appearance and Expressive Spatial Models for Human Pose Estimation , 2013, 2013 IEEE International Conference on Computer Vision.

[3]  Andrew Zisserman,et al.  Human Pose Estimation Using a Joint Pixel-wise and Part-wise Formulation , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[4]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[5]  Yi Li,et al.  Instance-Sensitive Fully Convolutional Networks , 2016, ECCV.

[6]  Bernt Schiele,et al.  2D Human Pose Estimation: New Benchmark and State of the Art Analysis , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[7]  Jonathan Tompson,et al.  Joint Training of a Convolutional Network and a Graphical Model for Human Pose Estimation , 2014, NIPS.

[8]  Rob Fergus,et al.  Depth Map Prediction from a Single Image using a Multi-Scale Deep Network , 2014, NIPS.