AUT LibraryAUT
View Item 
  •   Open Research
  • AUT Faculties
  • Faculty of Health and Environmental Sciences
  • School of Sport and Recreation
  • View Item
  •   Open Research
  • AUT Faculties
  • Faculty of Health and Environmental Sciences
  • School of Sport and Recreation
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Computational Intelligence for Qualitative Coaching Diagnostics: Automated Assessment of Tennis Swings to Improve Performance and Safety

Bačić, B; Hume, PA
Thumbnail
View/Open
Journal article (2.594Mb)
Permanent link
http://hdl.handle.net/10292/12738
Metadata
Show full metadata
Abstract
Coaching technology, wearables, and exergames can provide quantitative feedback based on measured activity, but there is little evidence of effective qualitative feedback to aid technique improvement. To achieve personalized qualitative feedback, we demonstrated a proof-of-concept prototype combining kinesiology and computational intelligence that could help improving tennis swing technique utilizing three-dimensional (3D) tennis motion data acquired from multicamera video. Expert data labeling relied on virtual 3D stick figure replay. Diverse assessment criteria for novice to those with intermediate skill levels and configurable coaching scenarios matched with a variety of tennis swings (22 backhands and 21 forehands), including good technique and common errors. A set of selected coaching rules (CRs) was transferred to adaptive assessment modules able to learn from data, evolve their internal structures, and produce autonomous personalized feedback, including verbal cues over virtual camera 3D replay and an end-of-session progress report. The prototype demonstrated autonomous assessment on future data based on learning from prior examples, aligned with skill level, flexible coaching scenarios, and CRs. The generated intuitive diagnostic feedback consisted of elements of safety and performance for tennis swing technique, where each swing sample was compared with the expert. For safety aspects of the relative swing width, the prototype showed improved assessment (from 81% to 91%) when taking into account occluded parts of the pelvis. This study has shown proof of concept for personalized qualitative feedback. The next generation of augmented coaching and exergaming systems will be able to help improve end user's sport discipline-specific techniques. By learning from small expert-labeled data sets, such systems will be able to adapt and provide personalized intuitive autonomous assessment and diagnostic feedback aligned with a specified coaching program and context requirements.
Keywords
Computational sport science; Pattern recognition; Sport technology; Human activities; Performance analysis; Sport analytics
Date
2018
Source
Big Data (2018), doi: 10.1089/big.2018.0062
Item Type
Journal Article
Publisher
Mary Ann Liebert, Inc.
DOI
10.1089/big.2018.0062
Publisher's Version
https://www.liebertpub.com/doi/10.1089/big.2018.0062
Rights Statement
Authors may archive their preprint manuscripts (version prior to peer review) at any time without restrictions.

Contact Us
  • Admin

Hosted by Tuwhera, an initiative of the Auckland University of Technology Library

 

 

Browse

Open ResearchTitlesAuthorsDateSchool of Sport and RecreationTitlesAuthorsDate

Alternative metrics

 

Statistics

For this itemFor all Open Research

Share

 
Follow @AUT_SC

Contact Us
  • Admin

Hosted by Tuwhera, an initiative of the Auckland University of Technology Library