Stereotyped gesture recognition: An analysis between HMM and SVM

Abstract

Stereotypic behaviours are present in both human and nonhuman primates. Usually, these behaviours are a welfare indicator. However, the stereotypic behaviours may be also a symptom of some mental disorder in the humans. A specific case is Autism Spectrum Disorder (ASD). The individuals with ASD may exhibit stereotypic behaviours through some gestures. The classic stereotyped gestures of autism are: (i) Body Rocking; (ii) Hand Flapping; and (iii) Top Spinning. This paper study the performance between two machine learning algorithms to recognition the stereotyped gestures typical of autism: (i) Hidden Markov Model [HMM]; and (ii) Support Vector Machine [SVM]. Sequence of orientations data from some joints obtained through a RGB-D (Red Green Blue - Depth) camera [Kinect®] are used for analysis. The results of these two machine learning algorithms are compared with state-of-the-art. The HMM approach proposed in this paper have shown 98.89% average recognition rate and 98.9% recall. This value is higher compared to the SVM approach and the others of art method presented.