Latent label mining for group activity recognition in basketball videos

Abstract Motion information has been widely exploited for group activity recognition in sports video. However, in order to model and extract the various motion information between the adjacent frames, existing algorithms only use the coarse video‐level labels as supervision cues. This may lead to th...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Lifang Wu, Zeyu Li, Ye Xiang, Meng Jian, Jialie Shen
Formato: article
Lenguaje:EN
Publicado: Wiley 2021
Materias:
Acceso en línea:https://doaj.org/article/631c0cbd2e244026b895fee7983080bd
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:631c0cbd2e244026b895fee7983080bd
record_format dspace
spelling oai:doaj.org-article:631c0cbd2e244026b895fee7983080bd2021-11-29T03:38:16ZLatent label mining for group activity recognition in basketball videos1751-96671751-965910.1049/ipr2.12265https://doaj.org/article/631c0cbd2e244026b895fee7983080bd2021-12-01T00:00:00Zhttps://doi.org/10.1049/ipr2.12265https://doaj.org/toc/1751-9659https://doaj.org/toc/1751-9667Abstract Motion information has been widely exploited for group activity recognition in sports video. However, in order to model and extract the various motion information between the adjacent frames, existing algorithms only use the coarse video‐level labels as supervision cues. This may lead to the ambiguity of extracted features and the omission of changing rules of motion patterns that are also important sports video recognition. In this paper, a latent label mining strategy for group activity recognition in basketball videos is proposed. The authors' novel strategy allows them to obtain the latent labels set for marking different frames in an unsupervised way, and build the frame‐level and video‐level representations with two separate levels of supervision signal. Firstly, the latent labels of motion patterns are digged using the unsupervised hierarchical clustering technique. The generated latent labels are then taken as the frame‐level supervision signal to train a deep CNN for the frame‐level features extraction. Lastly, the frame‐level features are fed into an LSTM network to build the spatio‐temporal representation for group activity recognition. Experimental results on the public NCAA dataset demonstrate that the proposed algorithm achieves state‐of‐the‐art performance.Lifang WuZeyu LiYe XiangMeng JianJialie ShenWileyarticlePhotographyTR1-1050Computer softwareQA76.75-76.765ENIET Image Processing, Vol 15, Iss 14, Pp 3487-3497 (2021)
institution DOAJ
collection DOAJ
language EN
topic Photography
TR1-1050
Computer software
QA76.75-76.765
spellingShingle Photography
TR1-1050
Computer software
QA76.75-76.765
Lifang Wu
Zeyu Li
Ye Xiang
Meng Jian
Jialie Shen
Latent label mining for group activity recognition in basketball videos
description Abstract Motion information has been widely exploited for group activity recognition in sports video. However, in order to model and extract the various motion information between the adjacent frames, existing algorithms only use the coarse video‐level labels as supervision cues. This may lead to the ambiguity of extracted features and the omission of changing rules of motion patterns that are also important sports video recognition. In this paper, a latent label mining strategy for group activity recognition in basketball videos is proposed. The authors' novel strategy allows them to obtain the latent labels set for marking different frames in an unsupervised way, and build the frame‐level and video‐level representations with two separate levels of supervision signal. Firstly, the latent labels of motion patterns are digged using the unsupervised hierarchical clustering technique. The generated latent labels are then taken as the frame‐level supervision signal to train a deep CNN for the frame‐level features extraction. Lastly, the frame‐level features are fed into an LSTM network to build the spatio‐temporal representation for group activity recognition. Experimental results on the public NCAA dataset demonstrate that the proposed algorithm achieves state‐of‐the‐art performance.
format article
author Lifang Wu
Zeyu Li
Ye Xiang
Meng Jian
Jialie Shen
author_facet Lifang Wu
Zeyu Li
Ye Xiang
Meng Jian
Jialie Shen
author_sort Lifang Wu
title Latent label mining for group activity recognition in basketball videos
title_short Latent label mining for group activity recognition in basketball videos
title_full Latent label mining for group activity recognition in basketball videos
title_fullStr Latent label mining for group activity recognition in basketball videos
title_full_unstemmed Latent label mining for group activity recognition in basketball videos
title_sort latent label mining for group activity recognition in basketball videos
publisher Wiley
publishDate 2021
url https://doaj.org/article/631c0cbd2e244026b895fee7983080bd
work_keys_str_mv AT lifangwu latentlabelminingforgroupactivityrecognitioninbasketballvideos
AT zeyuli latentlabelminingforgroupactivityrecognitioninbasketballvideos
AT yexiang latentlabelminingforgroupactivityrecognitioninbasketballvideos
AT mengjian latentlabelminingforgroupactivityrecognitioninbasketballvideos
AT jialieshen latentlabelminingforgroupactivityrecognitioninbasketballvideos
_version_ 1718407648716521472