Development of three dimensional foot behavior evaluation system based on AR technology

Abstract

Running form with pronation characteristic is an important key to choose suitable footwear and running gear. This paper describes development of a three-dimensional foot behavior measurement system with low-cost and small-footprint, and formulation of an evaluation parameter to identify runner's pronation characteristic accurately. The measurement system has SngleCamera3D (Photron Co., Ltd.) based on AR technology. In this system, subjects run on a treadmill with four AR markers attached on their heels and calves, while their running forms are recorded from behind by a single video camera. From a sequence of 2D images, three-dimensional orientation matrices of the heel and the calf during running are obtained by modified OpenCV image processing. We also proposed “pronation score” which took into account subtalar joint, heel inclination and tibial rotation angles during stance phase. As a result, the score showed good agreement with runner's pronation type evaluated by specialists. By using the proposed system, we evaluated 216 recreational runners in a major marathon event in Japan, and found that 22%, 40% and 38% of runners were underpronators, neutral pronators and overpronators, respectively.