Lower body pose estimation in team sports videos using Label-Grid classifier integrated with tracking-by-detection

Masaki Hayashi, Kyoko Oshima, Masamoto Tanabiki, Yoshimitsu Aoki

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

We propose a human lower body pose estimation method for team sport videos, which is integrated with tracking-by-detection technique. The proposed Label-Grid classifier uses the grid histogram feature of the tracked window from the tracker and estimates the lower body joint position of a specific joint as the class label of the multiclass classifiers, whose classes correspond to the candidate joint positions on the grid. By learning various types of player poses and scales of Histogram-of-Oriented Gradients features within one team sport, our method can estimate poses even if the players are motion-blurred and low-resolution images without requiring a motion-model regression or part-based model, which are popular vision-based human pose estimation techniques. Moreover, our method can estimate poses with part-occlusions and non-upright side poses, which part-detector-based methods find it difficult to estimate with only one model. Experimental results show the advantage of our method for side running poses and non-walking poses. The results also show the robustness of our method for a large variety of poses and scales in team sports videos.

Original languageEnglish
Pages (from-to)18-30
Number of pages13
JournalIPSJ Transactions on Computer Vision and Applications
Volume7
DOIs
Publication statusPublished - 2015

Fingerprint

Sports
Labels
Classifiers
Image resolution
Detectors

Keywords

  • Feature selection
  • Human pose estimation
  • People tracking
  • Random Forests
  • Tracking-by-detection

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Lower body pose estimation in team sports videos using Label-Grid classifier integrated with tracking-by-detection. / Hayashi, Masaki; Oshima, Kyoko; Tanabiki, Masamoto; Aoki, Yoshimitsu.

In: IPSJ Transactions on Computer Vision and Applications, Vol. 7, 2015, p. 18-30.

Research output: Contribution to journalArticle

@article{b0b7653e7ba94defb25075eac8d117f3,
title = "Lower body pose estimation in team sports videos using Label-Grid classifier integrated with tracking-by-detection",
abstract = "We propose a human lower body pose estimation method for team sport videos, which is integrated with tracking-by-detection technique. The proposed Label-Grid classifier uses the grid histogram feature of the tracked window from the tracker and estimates the lower body joint position of a specific joint as the class label of the multiclass classifiers, whose classes correspond to the candidate joint positions on the grid. By learning various types of player poses and scales of Histogram-of-Oriented Gradients features within one team sport, our method can estimate poses even if the players are motion-blurred and low-resolution images without requiring a motion-model regression or part-based model, which are popular vision-based human pose estimation techniques. Moreover, our method can estimate poses with part-occlusions and non-upright side poses, which part-detector-based methods find it difficult to estimate with only one model. Experimental results show the advantage of our method for side running poses and non-walking poses. The results also show the robustness of our method for a large variety of poses and scales in team sports videos.",
keywords = "Feature selection, Human pose estimation, People tracking, Random Forests, Tracking-by-detection",
author = "Masaki Hayashi and Kyoko Oshima and Masamoto Tanabiki and Yoshimitsu Aoki",
year = "2015",
doi = "10.2197/ipsjtcva.7.18",
language = "English",
volume = "7",
pages = "18--30",
journal = "IPSJ Transactions on Computer Vision and Applications",
issn = "1882-6695",
publisher = "Information Processing Society of Japan",

}

TY - JOUR

T1 - Lower body pose estimation in team sports videos using Label-Grid classifier integrated with tracking-by-detection

AU - Hayashi, Masaki

AU - Oshima, Kyoko

AU - Tanabiki, Masamoto

AU - Aoki, Yoshimitsu

PY - 2015

Y1 - 2015

N2 - We propose a human lower body pose estimation method for team sport videos, which is integrated with tracking-by-detection technique. The proposed Label-Grid classifier uses the grid histogram feature of the tracked window from the tracker and estimates the lower body joint position of a specific joint as the class label of the multiclass classifiers, whose classes correspond to the candidate joint positions on the grid. By learning various types of player poses and scales of Histogram-of-Oriented Gradients features within one team sport, our method can estimate poses even if the players are motion-blurred and low-resolution images without requiring a motion-model regression or part-based model, which are popular vision-based human pose estimation techniques. Moreover, our method can estimate poses with part-occlusions and non-upright side poses, which part-detector-based methods find it difficult to estimate with only one model. Experimental results show the advantage of our method for side running poses and non-walking poses. The results also show the robustness of our method for a large variety of poses and scales in team sports videos.

AB - We propose a human lower body pose estimation method for team sport videos, which is integrated with tracking-by-detection technique. The proposed Label-Grid classifier uses the grid histogram feature of the tracked window from the tracker and estimates the lower body joint position of a specific joint as the class label of the multiclass classifiers, whose classes correspond to the candidate joint positions on the grid. By learning various types of player poses and scales of Histogram-of-Oriented Gradients features within one team sport, our method can estimate poses even if the players are motion-blurred and low-resolution images without requiring a motion-model regression or part-based model, which are popular vision-based human pose estimation techniques. Moreover, our method can estimate poses with part-occlusions and non-upright side poses, which part-detector-based methods find it difficult to estimate with only one model. Experimental results show the advantage of our method for side running poses and non-walking poses. The results also show the robustness of our method for a large variety of poses and scales in team sports videos.

KW - Feature selection

KW - Human pose estimation

KW - People tracking

KW - Random Forests

KW - Tracking-by-detection

UR - http://www.scopus.com/inward/record.url?scp=84930028300&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84930028300&partnerID=8YFLogxK

U2 - 10.2197/ipsjtcva.7.18

DO - 10.2197/ipsjtcva.7.18

M3 - Article

VL - 7

SP - 18

EP - 30

JO - IPSJ Transactions on Computer Vision and Applications

JF - IPSJ Transactions on Computer Vision and Applications

SN - 1882-6695

ER -