Analysis of Children with Autism Spectrum Disorder (ASD) by Non-Verbal Language

Introduction

In this work, we develop an interactive multimedia system for the diagnosis of Autism Spectrum Disorder (ASD) of children. The system is composed of microphone array, optical sensors and a computer running own custom developed software. Children are allowed to play freely inside the specially designed room. 8 depth cameras and 4 high resolution RGB cameras are used to record multi-view behavior data. All sensors are aligned by camera calibration and multiple sensor fusion. Body skeleton, spatial position, gaze direction, head pose and gesture are extracted from the multi-view video streams. The interactive environment helps to improve the connection with reality in subjects having ASD. The system has been tested on a set of subjects. The effectiveness of the system has been recognized by a team of psychologists.

Demo

  • Interactive Diagnosis Chamber
Xiao Li (Shaw)
Xiao Li (Shaw)
Ph.D. Student

My research interests include computer vision, human-computer interaction, 3D vision and programmable matter.