Cargando…

3D Pose Detection of Closely Interactive Humans Using Multi-View Cameras

We propose a method to automatically detect 3D poses of closely interactive humans from sparse multi-view images at one time instance. It is a challenging problem due to the strong partial occlusion and truncation between humans and no tracking process to provide priori poses information. To solve t...

Descripción completa

Detalles Bibliográficos
Autores principales: Li, Xiu, Fan, Zhen, Liu, Yebin, Li, Yipeng, Dai, Qionghai
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2019
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6631029/
https://www.ncbi.nlm.nih.gov/pubmed/31242651
http://dx.doi.org/10.3390/s19122831
_version_ 1783435434575003648
author Li, Xiu
Fan, Zhen
Liu, Yebin
Li, Yipeng
Dai, Qionghai
author_facet Li, Xiu
Fan, Zhen
Liu, Yebin
Li, Yipeng
Dai, Qionghai
author_sort Li, Xiu
collection PubMed
description We propose a method to automatically detect 3D poses of closely interactive humans from sparse multi-view images at one time instance. It is a challenging problem due to the strong partial occlusion and truncation between humans and no tracking process to provide priori poses information. To solve this problem, we first obtain 2D joints in every image using OpenPose and human semantic segmentation results from Mask R-CNN. With the 3D joints triangulated from multi-view 2D joints, a two-stage assembling method is proposed to select the correct 3D pose from thousands of pose seeds combined by joint semantic meanings. We further present a novel approach to minimize the interpenetration between human shapes with close interactions. Finally, we test our method on multi-view human-human interaction (MHHI) datasets. Experimental results demonstrate that our method achieves high visualized correct rate and outperforms the existing method in accuracy and real-time capability.
format Online
Article
Text
id pubmed-6631029
institution National Center for Biotechnology Information
language English
publishDate 2019
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-66310292019-08-19 3D Pose Detection of Closely Interactive Humans Using Multi-View Cameras Li, Xiu Fan, Zhen Liu, Yebin Li, Yipeng Dai, Qionghai Sensors (Basel) Article We propose a method to automatically detect 3D poses of closely interactive humans from sparse multi-view images at one time instance. It is a challenging problem due to the strong partial occlusion and truncation between humans and no tracking process to provide priori poses information. To solve this problem, we first obtain 2D joints in every image using OpenPose and human semantic segmentation results from Mask R-CNN. With the 3D joints triangulated from multi-view 2D joints, a two-stage assembling method is proposed to select the correct 3D pose from thousands of pose seeds combined by joint semantic meanings. We further present a novel approach to minimize the interpenetration between human shapes with close interactions. Finally, we test our method on multi-view human-human interaction (MHHI) datasets. Experimental results demonstrate that our method achieves high visualized correct rate and outperforms the existing method in accuracy and real-time capability. MDPI 2019-06-25 /pmc/articles/PMC6631029/ /pubmed/31242651 http://dx.doi.org/10.3390/s19122831 Text en © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Li, Xiu
Fan, Zhen
Liu, Yebin
Li, Yipeng
Dai, Qionghai
3D Pose Detection of Closely Interactive Humans Using Multi-View Cameras
title 3D Pose Detection of Closely Interactive Humans Using Multi-View Cameras
title_full 3D Pose Detection of Closely Interactive Humans Using Multi-View Cameras
title_fullStr 3D Pose Detection of Closely Interactive Humans Using Multi-View Cameras
title_full_unstemmed 3D Pose Detection of Closely Interactive Humans Using Multi-View Cameras
title_short 3D Pose Detection of Closely Interactive Humans Using Multi-View Cameras
title_sort 3d pose detection of closely interactive humans using multi-view cameras
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6631029/
https://www.ncbi.nlm.nih.gov/pubmed/31242651
http://dx.doi.org/10.3390/s19122831
work_keys_str_mv AT lixiu 3dposedetectionofcloselyinteractivehumansusingmultiviewcameras
AT fanzhen 3dposedetectionofcloselyinteractivehumansusingmultiviewcameras
AT liuyebin 3dposedetectionofcloselyinteractivehumansusingmultiviewcameras
AT liyipeng 3dposedetectionofcloselyinteractivehumansusingmultiviewcameras
AT daiqionghai 3dposedetectionofcloselyinteractivehumansusingmultiviewcameras