Classification of a two-dimensional pose using a human skeleton
Marina M. Zhdanova, Oxana S. Balabaeva, Viacheslav Voronin, В. П. Федосов · MATEC Web of Conferences · 2017
This article proposes an approach for the human pose recognition based on to preliminary prepared high-level data of the human skeleton. The coordinates of the feature points on the human skeleton is constructed by using a neural network. In the presented method, the coordinates of the feature points are a normalized relative of the human body height and the centre of gravity. We use these data for training the neural network.