Development of an Automated Motion Evaluation System from Wearable Sensor Devices for Ski Jumping

Heike Brock, Yuji Ohgi, Kazuya Seo

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

We developed a system for the automatic evaluation of ski jumps on the base of machine learning algorithms. In several capture sessions during summer jump season, motions of four junior ski jumpers were captured by inertial sensors over the complete jump, from the start of the in-run to the end of the outrun phase. Additionally, style points were collected from an experienced judge to serve as ground truth and control data for the machine learning algorithms. All jumps were randomly separated into a training and a test database. Next, we determined kinematic factors such as body segment orientation and body joint position and segmented every jump into its main motion phases on the base of the raw sensor data. Specific motion characteristics that influence the performance and length of a jump were then used to create machine style knowledge from the training database in compliance to the official scoring guidelines. In a last step, we computed the similarity of every jump within the test database to the trained knowledge on a ski jump's style parameters and evaluated the presence or absence of jump errors. Results showed that the computed error annotations conform largely to the human-based judging scores. Adding automatic score measures to the algorithm in future, this method might be an important step towards better measurability and objectivity of performance-oriented sports in form of a mobile competition evaluation system.

Original languageEnglish
Pages (from-to)694-699
Number of pages6
JournalUnknown Journal
Volume147
DOIs
Publication statusPublished - 2016

Fingerprint

Ski jumps
Jump
Databases
Equipment and Supplies
Sensor
Learning algorithms
Learning systems
Motion
Evaluation
evaluation
Athletic Performance
Sensors
objectivity
Sports
Biomechanical Phenomena
learning
performance
Kinematics
Joints
Guidelines

Keywords

  • machine learning
  • Motion analysis
  • motion evaluation
  • ski jumping
  • wearable sensor devices

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Development of an Automated Motion Evaluation System from Wearable Sensor Devices for Ski Jumping. / Brock, Heike; Ohgi, Yuji; Seo, Kazuya.

In: Unknown Journal, Vol. 147, 2016, p. 694-699.

Research output: Contribution to journalArticle

@article{b83c01f62c2b41508e54a29fa4ac0ca0,
title = "Development of an Automated Motion Evaluation System from Wearable Sensor Devices for Ski Jumping",
abstract = "We developed a system for the automatic evaluation of ski jumps on the base of machine learning algorithms. In several capture sessions during summer jump season, motions of four junior ski jumpers were captured by inertial sensors over the complete jump, from the start of the in-run to the end of the outrun phase. Additionally, style points were collected from an experienced judge to serve as ground truth and control data for the machine learning algorithms. All jumps were randomly separated into a training and a test database. Next, we determined kinematic factors such as body segment orientation and body joint position and segmented every jump into its main motion phases on the base of the raw sensor data. Specific motion characteristics that influence the performance and length of a jump were then used to create machine style knowledge from the training database in compliance to the official scoring guidelines. In a last step, we computed the similarity of every jump within the test database to the trained knowledge on a ski jump's style parameters and evaluated the presence or absence of jump errors. Results showed that the computed error annotations conform largely to the human-based judging scores. Adding automatic score measures to the algorithm in future, this method might be an important step towards better measurability and objectivity of performance-oriented sports in form of a mobile competition evaluation system.",
keywords = "machine learning, Motion analysis, motion evaluation, ski jumping, wearable sensor devices",
author = "Heike Brock and Yuji Ohgi and Kazuya Seo",
year = "2016",
doi = "10.1016/j.proeng.2016.06.248",
language = "English",
volume = "147",
pages = "694--699",
journal = "Mathematical Social Sciences",
issn = "0165-4896",
publisher = "Elsevier",

}

TY - JOUR

T1 - Development of an Automated Motion Evaluation System from Wearable Sensor Devices for Ski Jumping

AU - Brock, Heike

AU - Ohgi, Yuji

AU - Seo, Kazuya

PY - 2016

Y1 - 2016

N2 - We developed a system for the automatic evaluation of ski jumps on the base of machine learning algorithms. In several capture sessions during summer jump season, motions of four junior ski jumpers were captured by inertial sensors over the complete jump, from the start of the in-run to the end of the outrun phase. Additionally, style points were collected from an experienced judge to serve as ground truth and control data for the machine learning algorithms. All jumps were randomly separated into a training and a test database. Next, we determined kinematic factors such as body segment orientation and body joint position and segmented every jump into its main motion phases on the base of the raw sensor data. Specific motion characteristics that influence the performance and length of a jump were then used to create machine style knowledge from the training database in compliance to the official scoring guidelines. In a last step, we computed the similarity of every jump within the test database to the trained knowledge on a ski jump's style parameters and evaluated the presence or absence of jump errors. Results showed that the computed error annotations conform largely to the human-based judging scores. Adding automatic score measures to the algorithm in future, this method might be an important step towards better measurability and objectivity of performance-oriented sports in form of a mobile competition evaluation system.

AB - We developed a system for the automatic evaluation of ski jumps on the base of machine learning algorithms. In several capture sessions during summer jump season, motions of four junior ski jumpers were captured by inertial sensors over the complete jump, from the start of the in-run to the end of the outrun phase. Additionally, style points were collected from an experienced judge to serve as ground truth and control data for the machine learning algorithms. All jumps were randomly separated into a training and a test database. Next, we determined kinematic factors such as body segment orientation and body joint position and segmented every jump into its main motion phases on the base of the raw sensor data. Specific motion characteristics that influence the performance and length of a jump were then used to create machine style knowledge from the training database in compliance to the official scoring guidelines. In a last step, we computed the similarity of every jump within the test database to the trained knowledge on a ski jump's style parameters and evaluated the presence or absence of jump errors. Results showed that the computed error annotations conform largely to the human-based judging scores. Adding automatic score measures to the algorithm in future, this method might be an important step towards better measurability and objectivity of performance-oriented sports in form of a mobile competition evaluation system.

KW - machine learning

KW - Motion analysis

KW - motion evaluation

KW - ski jumping

KW - wearable sensor devices

UR - http://www.scopus.com/inward/record.url?scp=84982899535&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84982899535&partnerID=8YFLogxK

U2 - 10.1016/j.proeng.2016.06.248

DO - 10.1016/j.proeng.2016.06.248

M3 - Article

AN - SCOPUS:84982899535

VL - 147

SP - 694

EP - 699

JO - Mathematical Social Sciences

JF - Mathematical Social Sciences

SN - 0165-4896

ER -