A Computer Vision Based Approach for Understanding Emotional Involvements in Children with Autism Spectrum Disorders

It has been proved that Autism Spectrum Disorders (ASD) are associated with amplified emotional responses and poor emotional control. Underlying mechanisms and characteristics of these difficulties in using, sharing and responding to emotions are still not understood. This is because advanced computational approaches for studying details of facial expressions have been based on the use of invasive instruments (such markers for motion capture or Electromyographs) that can affect the behaviors and, above all, restrict the possibility to implement diagnostic and evaluation tools. Recent non-invasive technological frameworks based on computer vision can be applied to overcome this knowledge gap and this paper is right aimed at demonstrating how facial measurements from images can be exploited to compare how ASD children react to external stimuli with respect a control set of children. This paper has a double layer of contribution: on the one hand it aims at proposing the use of a single-camera system for facial expression analysis and, on the other hand, it presents a study on how extracted facial data could be used to analyze how the overall and local facial dynamics of children with ASD differ from their typically developing peers. In other words, this study explores the feasibility of the introduction of numerical approaches for the diagnosis and evaluation of autistic spectrum disorders in preschool children.