Näytä suppeat kuvailutiedot

dc.contributor.authorCronin, Neil
dc.contributor.authorRantalainen, Timo
dc.contributor.authorAhtiainen, Juha
dc.contributor.authorHynynen, Esa
dc.contributor.authorWaller, Benjamin
dc.date.accessioned2019-05-08T08:06:41Z
dc.date.available2020-04-19T21:35:09Z
dc.date.issued2019
dc.identifier.citationCronin, N., Rantalainen, T., Ahtiainen, J., Hynynen, E., & Waller, B. (2019). Markerless 2D kinematic analysis of underwater running : A deep learning approach. <i>Journal of Biomechanics</i>, <i>87</i>, 75-82. <a href="https://doi.org/10.1016/j.jbiomech.2019.02.021" target="_blank">https://doi.org/10.1016/j.jbiomech.2019.02.021</a>
dc.identifier.otherCONVID_28956280
dc.identifier.otherTUTKAID_80873
dc.identifier.urihttps://jyx.jyu.fi/handle/123456789/63811
dc.description.abstractKinematic analysis is often performed with a camera system combined with reflective markers placed over bony landmarks. This method is restrictive (and often expensive), and limits the ability to perform analyses outside of the lab. In the present study, we used a markerless deep learning-based method to perform 2D kinematic analysis of deep water running, a task that poses several challenges to image processing methods. A single GoPro camera recorded sagittal plane lower limb motion. A deep neural network was trained using data from 17 individuals, and then used to predict the locations of markers that approximated joint centres. We found that 300–400 labelled images were sufficient to train the network to be able to position joint markers with an accuracy similar to that of a human labeler (mean difference < 3 pixels, around 1 cm). This level of accuracy is sufficient for many 2D applications, such as sports biomechanics, coaching/training, and rehabilitation. The method was sensitive enough to differentiate between closely-spaced running cadences (45–85 strides per minute in increments of 5). We also found high test–retest reliability of mean stride data, with between-session correlation coefficients of 0.90–0.97. Our approach represents a low-cost, adaptable solution for kinematic analysis, and could easily be modified for use in other movements and settings. Using additional cameras, this approach could also be used to perform 3D analyses. The method presented here may have broad applications in different fields, for example by enabling markerless motion analysis to be performed during rehabilitation, training or even competition environments.fi
dc.format.mimetypeapplication/pdf
dc.language.isoeng
dc.publisherPergamon Press
dc.relation.ispartofseriesJournal of Biomechanics
dc.rightsCC BY-NC-ND 4.0
dc.subject.otherdeep water running
dc.subject.otherdeep learning
dc.titleMarkerless 2D kinematic analysis of underwater running : A deep learning approach
dc.typearticle
dc.identifier.urnURN:NBN:fi:jyu-201905072424
dc.contributor.laitosLiikuntatieteellinen tiedekuntafi
dc.contributor.laitosFaculty of Sport and Health Sciencesen
dc.contributor.oppiaineBiomekaniikkafi
dc.contributor.oppiaineValmennus- ja testausoppifi
dc.contributor.oppiaineBiomechanicsen
dc.contributor.oppiaineScience of Sport Coaching and Fitness Testingen
dc.type.urihttp://purl.org/eprint/type/JournalArticle
dc.date.updated2019-05-07T09:15:14Z
dc.type.coarhttp://purl.org/coar/resource_type/c_2df8fbb1
dc.description.reviewstatuspeerReviewed
dc.format.pagerange75-82
dc.relation.issn0021-9290
dc.relation.numberinseries0
dc.relation.volume87
dc.type.versionacceptedVersion
dc.rights.copyright© 2019 Elsevier Ltd.
dc.rights.accesslevelopenAccessfi
dc.subject.ysovesijuoksu
dc.subject.ysoliikeoppi
dc.subject.ysotekoäly
dc.subject.ysoliikeanalyysi
dc.format.contentfulltext
jyx.subject.urihttp://www.yso.fi/onto/yso/p2238
jyx.subject.urihttp://www.yso.fi/onto/yso/p16028
jyx.subject.urihttp://www.yso.fi/onto/yso/p2616
jyx.subject.urihttp://www.yso.fi/onto/yso/p24952
dc.rights.urlhttps://creativecommons.org/licenses/by-nc-nd/4.0/
dc.relation.doi10.1016/j.jbiomech.2019.02.021
dc.type.okmA1


Aineistoon kuuluvat tiedostot

Thumbnail

Aineisto kuuluu seuraaviin kokoelmiin

Näytä suppeat kuvailutiedot

CC BY-NC-ND 4.0
Ellei muuten mainita, aineiston lisenssi on CC BY-NC-ND 4.0