Abstract—In recent years, motion estimation analysis has become one of the vital research areas in sport and has attracted the interest of many researchers toward events such as swimming, pole vaulting, and hurdling. In this paper, we present a novel method for determining the step length, speed, and the feet-contact-time on the running track of a pole vault athlete using a mono-camera arrangement. The step length and step frequency are essential descriptors of the approach run in pole vaulting. The approach along a linear trajectory is familiar to many throwing and jumping events. The measurement setting and image processing, as well as the step registration stages such as the block matching and optimal flow algorithm are presented and compared to alternative methods. The validation of the step size and step frequency accuracy is provided, using manually digitized step sizes as the baseline. The proposed methodology is efficient and straightforward, providing immediate feedback to the athlete and coaches. We were also successful in building a basic Graphical User Interface (GUI) to illustrate pole-vaulting actions during a performance. This research could be used as an initial step for developing a fully interactive platform that is capable of yielding supportive instructions to the athletes and the coaches on a real-time basis for self-assessment and further improvement.

Index Terms—motion estimation, gait recognition, graphical user interface, computer vision, pole vaulter action

Cite: Rajeev Kanth, Jukka-Pekka Skön, Kari Lehtomäki, Paavo Nevalainen, Petra Virjonen, Mikko-Jussi Laakso, and Jukka Heikkonen, "Image Analysis and Development of Graphical User Interface for Pole Vault Action," Journal of Image and Graphics, Vol. 6, No. 2, pp. 109-116, December 2018. doi: 10.18178/joig.6.2.109-116