A Computer Vision-Based Physical Activity Application for Children with Autism


Obuz E., Toraman K., Solak S.

2025 7th International Congress on Human-Computer Interaction, Optimization and Robotic Applications (ICHORA), Ankara, Turkey, 23 - 24 May 2025, pp.1-4, (Full Text)

  • Publication Type: Conference Paper / Full Text
  • City: Ankara
  • Country: Turkey
  • Page Numbers: pp.1-4
  • Kocaeli University Affiliated: Yes

Abstract

Autism Spectrum Disorder (ASD) is a neurological condition characterized by challenges in social interaction, communication, repetitive behaviors, and restricted interests. While the exact cause remains unknown, ASD results from genetic and environmental factors and persists throughout life. Early diagnosis and intervention enhance quality of life and potential. Games have proven effective in nurturing socialization, skill acquisition, and adaptability in children with ASD, leveraging their structured and visually appealing design.This study adopted a game-based approach to support motor skills, socialization, and learning in children with ASD. Using computer vision techniques, children’s movements were captured via cameras and transformed into interactive virtual characters in real time. Activities included movement mimicry games to improve attention and motor skills, dance games for physical coordination, and turn-based games to enhance social skills. The system, developed in this study using Google MediaPipe Pose and Unity, is cost-effective and suitable for preschools and home use. Feedback from educators at the Autism Sports and Life Center guided system refinement. The work presented in this article offers a robust foundation for fostering independence and social participation in children with ASD.