This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ PrePrints) and either DOI or URL of the article must be cited.
Emotion recognition can improve the quality of patient care, product development and human-machine interaction.Psychological studies indicate that emotional state can be expressed in the way people walk,and the human gait can be used to reveal a person's emotional state.This paper proposes a novel method to do emotion recognition by using Microsoft Kinect to record gait patterns and train machine learning algorithms for emotion recognition. 59 subjects are recruited, and their gait patterns are recorded by two Kinect cameras.Joint selection, coordinate system transformation, sliding window gauss filtering,differential operation, and data segmentation are used for data preprocessing.We run Fourier transformation to extract features from the gait patterns and utilize Principal Component Analysis(PCA) for feature selection. By using NaiveBayes, RandomForests, LibSVM and SMO classifiers, the accuracy of recognition between natural and angry emotions can reach 80%,and the accuracy of recognition between natural and happy emotions can reach above 70%.The result indicates that Kinect can be used in the recognition of emotions with fairly well performance.
This version contains a new file.
Figure 1:The description of the experiment environment