A LSTM-based Approach for Gait Emotion Recognition

Gait Emotion Recognition (GER) is a popular problem which has applications in a variety of fields, including smart home design, cognitive systems, border security, robotics, virtual reality, and gaming. In the recent years, several Deep Learning (DL) based approaches for GER have been adopted. However, a vast majority of such methods rely on Deep Neural Networks (DNNs) with significant number of model parameters which are neither robust, nor efficient. This paper contributes to the domain of knowledge by presenting a novel light architecture for inferring human emotions through gait. It outperforms all recent deep learning methods, while having the lowest inference time for each gait sample.