Flexible top-view human pose estimation for detection system via CNN

Ryuji Go, Yoshimitsu Aoki

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We propose the DeepPose-based pose estimation system that is flexible with the change of bounding-box range for top-view images. Our purpose is to link person detection system and pose estimation system. We introduce Bounding-box Curriculum Learning (BCL) and Recurrent Pose Estimation (RPE). BCL is a learning technique of CNN inspired from Curriculum Learning. RPE is a recurrent process of pose estimation that fixes the bounding-box range in response to the estimated results. We show the effect of proposed methods compared to normal learned CNN-based pose estimator on our original top-view dataset.

Original languageEnglish
Title of host publication2016 IEEE 5th Global Conference on Consumer Electronics, GCCE 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509023332
DOIs
Publication statusPublished - 2016 Dec 27
Event5th IEEE Global Conference on Consumer Electronics, GCCE 2016 - Kyoto, Japan
Duration: 2016 Oct 112016 Oct 14

Other

Other5th IEEE Global Conference on Consumer Electronics, GCCE 2016
CountryJapan
CityKyoto
Period16/10/1116/10/14

Fingerprint

learning
boxes
Curricula
estimators
fixing

Keywords

  • Convolutional Neural Networks
  • Pose Estimation
  • Top-view

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Computer Science Applications
  • Hardware and Architecture
  • Instrumentation

Cite this

Go, R., & Aoki, Y. (2016). Flexible top-view human pose estimation for detection system via CNN. In 2016 IEEE 5th Global Conference on Consumer Electronics, GCCE 2016 [7800406] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/GCCE.2016.7800406

Flexible top-view human pose estimation for detection system via CNN. / Go, Ryuji; Aoki, Yoshimitsu.

2016 IEEE 5th Global Conference on Consumer Electronics, GCCE 2016. Institute of Electrical and Electronics Engineers Inc., 2016. 7800406.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Go, R & Aoki, Y 2016, Flexible top-view human pose estimation for detection system via CNN. in 2016 IEEE 5th Global Conference on Consumer Electronics, GCCE 2016., 7800406, Institute of Electrical and Electronics Engineers Inc., 5th IEEE Global Conference on Consumer Electronics, GCCE 2016, Kyoto, Japan, 16/10/11. https://doi.org/10.1109/GCCE.2016.7800406
Go R, Aoki Y. Flexible top-view human pose estimation for detection system via CNN. In 2016 IEEE 5th Global Conference on Consumer Electronics, GCCE 2016. Institute of Electrical and Electronics Engineers Inc. 2016. 7800406 https://doi.org/10.1109/GCCE.2016.7800406
Go, Ryuji ; Aoki, Yoshimitsu. / Flexible top-view human pose estimation for detection system via CNN. 2016 IEEE 5th Global Conference on Consumer Electronics, GCCE 2016. Institute of Electrical and Electronics Engineers Inc., 2016.
@inproceedings{9fff84fa952340e7b14358943cd433cb,
title = "Flexible top-view human pose estimation for detection system via CNN",
abstract = "We propose the DeepPose-based pose estimation system that is flexible with the change of bounding-box range for top-view images. Our purpose is to link person detection system and pose estimation system. We introduce Bounding-box Curriculum Learning (BCL) and Recurrent Pose Estimation (RPE). BCL is a learning technique of CNN inspired from Curriculum Learning. RPE is a recurrent process of pose estimation that fixes the bounding-box range in response to the estimated results. We show the effect of proposed methods compared to normal learned CNN-based pose estimator on our original top-view dataset.",
keywords = "Convolutional Neural Networks, Pose Estimation, Top-view",
author = "Ryuji Go and Yoshimitsu Aoki",
year = "2016",
month = "12",
day = "27",
doi = "10.1109/GCCE.2016.7800406",
language = "English",
booktitle = "2016 IEEE 5th Global Conference on Consumer Electronics, GCCE 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Flexible top-view human pose estimation for detection system via CNN

AU - Go, Ryuji

AU - Aoki, Yoshimitsu

PY - 2016/12/27

Y1 - 2016/12/27

N2 - We propose the DeepPose-based pose estimation system that is flexible with the change of bounding-box range for top-view images. Our purpose is to link person detection system and pose estimation system. We introduce Bounding-box Curriculum Learning (BCL) and Recurrent Pose Estimation (RPE). BCL is a learning technique of CNN inspired from Curriculum Learning. RPE is a recurrent process of pose estimation that fixes the bounding-box range in response to the estimated results. We show the effect of proposed methods compared to normal learned CNN-based pose estimator on our original top-view dataset.

AB - We propose the DeepPose-based pose estimation system that is flexible with the change of bounding-box range for top-view images. Our purpose is to link person detection system and pose estimation system. We introduce Bounding-box Curriculum Learning (BCL) and Recurrent Pose Estimation (RPE). BCL is a learning technique of CNN inspired from Curriculum Learning. RPE is a recurrent process of pose estimation that fixes the bounding-box range in response to the estimated results. We show the effect of proposed methods compared to normal learned CNN-based pose estimator on our original top-view dataset.

KW - Convolutional Neural Networks

KW - Pose Estimation

KW - Top-view

UR - http://www.scopus.com/inward/record.url?scp=85010302965&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85010302965&partnerID=8YFLogxK

U2 - 10.1109/GCCE.2016.7800406

DO - 10.1109/GCCE.2016.7800406

M3 - Conference contribution

AN - SCOPUS:85010302965

BT - 2016 IEEE 5th Global Conference on Consumer Electronics, GCCE 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -