Please use this identifier to cite or link to this item:
https://doi.org/10.48548/pubdata-1539
| Resource type | Journal Article |
| Title(s) | Toward Automatically Labeling Situations in Soccer |
| DOI | 10.48548/pubdata-1539 |
| Handle | 20.500.14123/1615 |
| Creator | Fassmeyer, Dennis  0009-0003-9330-0992Anzer, Gabriel 0000-0003-3129-8359Bauer, Pascal 0000-0001-8613-6635Brefeld, Ulf 0000-0001-9600-6463 |
| Abstract | We study the automatic annotation of situations in soccer games. At first sight, this translates nicely into a standard supervised learning problem. However, in a fully supervised setting, predictive accuracies are supposed to correlate positively with the amount of labeled situations: more labeled training data simply promise better performance. Unfortunately, non-trivially annotated situations in soccer games are scarce, expensive and almost always require human experts; a fully supervised approach appears infeasible. Hence, we split the problem into two parts and learn (i) a meaningful feature representation using variational autoencoders on unlabeled data at large scales and (ii) a large-margin classifier acting in this feature space but utilize only a few (manually) annotated examples of the situation of interest. We propose four different architectures of the variational autoencoder and empirically study the detection of corner kicks, crosses and counterattacks. We observe high predictive accuracies above 90% AUC irrespectively of the task. |
| Language | English |
| Keywords | Sports Analytics; Soccer; Data Tracking; Variational Autoencoders |
| Year of publication in PubData | 2024 |
| Publishing type | Parallel publication |
| Publication version | Published version |
| Date issued | 2021-11-03 |
| Creation context | Research |
| Notes | This publication was funded by the Open Access Publication Fund of Leuphana University Lüneburg. |
| Published by | Medien- und Informationszentrum, Leuphana Universität Lüneburg |
| Related resources |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
Fassmeyer_Toward_Automatically_Labeling_Situations_in_Soccer.pdf License: 
open-access | 1.95 MB | Adobe PDF | View/Open |
Items in PubData are protected by copyright, with all rights reserved, unless otherwise indicated.
Views
BibTeX
RIS
Item Export Bar
Datacite XML
OpenAIRE4
Access statistics
Page view(s): 26
Download(s): 1