Accessibility Improvement of Leisure Sports “Mölkky” for Visually Impaired Players Using AI Vision

Abstract

To establish an accessible sport that blind and low vision players can easily play with sighted players, we focused on the Finnish leisure sport, Mölkky. Trying to play it with blind and low vision players revealed that primitive assisting way such as clapping hands or explaining positions by voice were useful to some extent. On the other hand, the fact is observed that the throwing form of a congenital blind player was apparently different from the forms of other players. Therefore, to show that difference to the blind player, a prototype system that consists of a computer and a handy size wireless haptic device equipped with servo motors and a model arm was developed. The system detects the positions of a shoulder, an elbow, and a wrist on the throwing arm using a Mediapipe framework provided by Google. Then, a shoulder angle and an elbow angle are calculated by these positions, and the computer send these data to the haptic device via Bluetooth. Finally, the blind player touches the device and understands the motion of the throwing arm. The system was assessed by three visually impaired people, and they reported that they could clearly recognize the difference between throwing motions.