Image-based center of mass estimation of the human body via 3D shape and kinematic structure

Tomoya Kaichi, Shohei Mori, Hideo Saito, Kosuke Takahashi, Dan Mikami, Mariko Isogawa, Yoshinori Kusachi

Research output: Contribution to journalArticleResearchpeer-review

Abstract

This paper presents a method to estimate a time-sequential trajectory of the center of mass (CoM) of an athlete from a multi-view set of cameras. Collecting the CoM typically requires large-scale measuring systems or attaching sensors to the athletes. To mitigate such hardware limitations, the present study takes a multi-view video-based approach. The proposed method reconstructs subjects’ voxels from a set of multi-view frames and weights each voxel with body part-dependent weights to calculate a CoM. Our results, using real data measured in a studio, showed that the proposed method can estimate CoM within 20 mm concerning center of pressure measures.
Original languageEnglish
Article number17
Number of pages9
JournalSports Engineering
Volume22
DOIs
Publication statusE-pub ahead of print - 2019

Fingerprint

Studios
Kinematics
Cameras
Trajectories
Hardware
Sensors

Cite this

Image-based center of mass estimation of the human body via 3D shape and kinematic structure. / Kaichi, Tomoya; Mori, Shohei; Saito, Hideo; Takahashi, Kosuke; Mikami, Dan; Isogawa, Mariko; Kusachi, Yoshinori.

In: Sports Engineering, Vol. 22, 17, 2019.

Research output: Contribution to journalArticleResearchpeer-review

Kaichi, Tomoya ; Mori, Shohei ; Saito, Hideo ; Takahashi, Kosuke ; Mikami, Dan ; Isogawa, Mariko ; Kusachi, Yoshinori. / Image-based center of mass estimation of the human body via 3D shape and kinematic structure. In: Sports Engineering. 2019 ; Vol. 22.
@article{b7d32e35f6634616b3f6d4f8394062f5,
title = "Image-based center of mass estimation of the human body via 3D shape and kinematic structure",
abstract = "This paper presents a method to estimate a time-sequential trajectory of the center of mass (CoM) of an athlete from a multi-view set of cameras. Collecting the CoM typically requires large-scale measuring systems or attaching sensors to the athletes. To mitigate such hardware limitations, the present study takes a multi-view video-based approach. The proposed method reconstructs subjects’ voxels from a set of multi-view frames and weights each voxel with body part-dependent weights to calculate a CoM. Our results, using real data measured in a studio, showed that the proposed method can estimate CoM within 20 mm concerning center of pressure measures.",
author = "Tomoya Kaichi and Shohei Mori and Hideo Saito and Kosuke Takahashi and Dan Mikami and Mariko Isogawa and Yoshinori Kusachi",
year = "2019",
doi = "10.1007/s12283-019-0309-2",
language = "English",
volume = "22",
journal = "Sports Engineering",
issn = "1369-7072",
publisher = "Springer",

}

TY - JOUR

T1 - Image-based center of mass estimation of the human body via 3D shape and kinematic structure

AU - Kaichi, Tomoya

AU - Mori, Shohei

AU - Saito, Hideo

AU - Takahashi, Kosuke

AU - Mikami, Dan

AU - Isogawa, Mariko

AU - Kusachi, Yoshinori

PY - 2019

Y1 - 2019

N2 - This paper presents a method to estimate a time-sequential trajectory of the center of mass (CoM) of an athlete from a multi-view set of cameras. Collecting the CoM typically requires large-scale measuring systems or attaching sensors to the athletes. To mitigate such hardware limitations, the present study takes a multi-view video-based approach. The proposed method reconstructs subjects’ voxels from a set of multi-view frames and weights each voxel with body part-dependent weights to calculate a CoM. Our results, using real data measured in a studio, showed that the proposed method can estimate CoM within 20 mm concerning center of pressure measures.

AB - This paper presents a method to estimate a time-sequential trajectory of the center of mass (CoM) of an athlete from a multi-view set of cameras. Collecting the CoM typically requires large-scale measuring systems or attaching sensors to the athletes. To mitigate such hardware limitations, the present study takes a multi-view video-based approach. The proposed method reconstructs subjects’ voxels from a set of multi-view frames and weights each voxel with body part-dependent weights to calculate a CoM. Our results, using real data measured in a studio, showed that the proposed method can estimate CoM within 20 mm concerning center of pressure measures.

U2 - 10.1007/s12283-019-0309-2

DO - 10.1007/s12283-019-0309-2

M3 - Article

VL - 22

JO - Sports Engineering

JF - Sports Engineering

SN - 1369-7072

M1 - 17

ER -