Running Video Matching Algorithm Focusing on the Periodicity and Linearity of Running Motion

Yoshiteru Yamamoto, Keisuke Doman, Yoshiya Hotta, Yoshito Mekada

Abstract

This paper proposes a framework for frame-by-frame matching of a performance video pair focusing on the periodicity and the linearity of running motion toward running form analysis support. The proposed method is composed of four steps: 1) 2D pose estimation, 2) correspondent phase detection, 3) linear matching, and 4) expansion matching. In the experiments, the proposed method could accurately match performance videos for different runners compared with the previous method. We confirmed the effectiveness and the usefulness of the proposed method.

Download


Paper Citation


in Harvard Style

Yamamoto Y., Doman K., Hotta Y. and Mekada Y. (2018). Running Video Matching Algorithm Focusing on the Periodicity and Linearity of Running Motion.In IcSPORTS 2018 - Extended Abstracts - Volume 1: icSPORTS, ISBN , pages 24-28


in Bibtex Style

@conference{icsports18,
author={Yoshiteru Yamamoto and Keisuke Doman and Yoshiya Hotta and Yoshito Mekada},
title={Running Video Matching Algorithm Focusing on the Periodicity and Linearity of Running Motion},
booktitle={IcSPORTS 2018 - Extended Abstracts - Volume 1: icSPORTS,},
year={2018},
pages={24-28},
publisher={SciTePress},
organization={INSTICC},
doi={},
isbn={},
}


in EndNote Style

TY - CONF

JO - IcSPORTS 2018 - Extended Abstracts - Volume 1: icSPORTS,
TI - Running Video Matching Algorithm Focusing on the Periodicity and Linearity of Running Motion
SN -
AU - Yamamoto Y.
AU - Doman K.
AU - Hotta Y.
AU - Mekada Y.
PY - 2018
SP - 24
EP - 28
DO -