kth.sePublications KTH
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Exploring Feature Extraction for Event Tracking in Beach Volleyball: Spatiotemporal Feature Extraction using Human Pose Estimation
KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Health Informatics and Logistics.
KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Health Informatics and Logistics.
2025 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
Utforskan av Feature Extraction för händelsespårning inom Beach Volleyboll : Spatiotemporal Feature Extraction med hjälp av Human Pose Estimation (Swedish)
Abstract [en]

This study explores extracting features from Beach Volleyball video data using Human Pose Estimation models and evaluating the features by training and assessing a Machine Learning model with the aforementioned features. The methodology of the study consists of a literature study, interviews, pre-experiments to help decide on a pipeline and, extracting and evaluating features. The results of the study show that the models trained on the extraced features did not generalise well on unseen data. This could be due to the training dataset being too small for the choice of ML model orthe choice of evaluation method.

Abstract [sv]

Denna studie utforskar möjligheten att utföra feature extraction på Beach Volleyboll video data genom att använda Human Pose Estimation modeller och utvärderar dessa features genom att träna och analysera maskininlärningsmodeller med dem. Metoden består av en litteraturstudie, intervjuer, förexperiment för att stödja valet av verktyg till pipelinen, och extraherandet samt utvärderingen av de features som extraherades. Resultaten visar att modellerna som tränades på dessa extraherade features inte generaliserade bra på nytt data vilket kan bero på att träningsdatasetet var för litet i storlek för vald ML model eller valet av utvärderingsmetod.

Place, publisher, year, edition, pages
2025.
Series
TRITA-CBH-GRU ; 2025:108
Keywords [en]
machine learning, feature extraction, sports technology, sports analysis, pose estimation, object detection, human activity recognition, beach volleyball
Keywords [sv]
maskin inlärning, feature extraction, sportteknologi, sportanalys, pose estimation, object detection, human activity recognition, beach volleyboll
National Category
Artificial Intelligence
Identifiers
URN: urn:nbn:se:kth:diva-364385OAI: oai:DiVA.org:kth-364385DiVA, id: diva2:1968067
Subject / course
Computer Technology, Program- and System Development
Educational program
Bachelor of Science in Engineering - Computer Engineering
Supervisors
Examiners
Available from: 2025-06-12 Created: 2025-06-12 Last updated: 2025-06-12Bibliographically approved

Open Access in DiVA

Exploring Feature Extraction for Event Tracking in Beach Volleyball(1320 kB)277 downloads
File information
File name FULLTEXT01.pdfFile size 1320 kBChecksum SHA-512
4149aaba3cc42bf6a7721c3634e832adf22aa6f16b2a9b3a1599d0a2d2c2ae37c570ab443c3c52f290ced61cd334894c38b29858ad8e3efc00b7121da152d052
Type fulltextMimetype application/pdf

By organisation
Health Informatics and Logistics
Artificial Intelligence

Search outside of DiVA

GoogleGoogle Scholar
Total: 278 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 355 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf