Joint Probabilistic People Detection in Overlapping Depth Images
Privacy-preserving high-quality people detection is a vital computer vision task for various indoor scenarios, e.g. people counting, customer behavior analysis, ambient assisted living or smart homes. In this work a novel approach for people detection in multiple overlapping depth images is proposed...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2020-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/8985332/ |
id |
doaj-a1d513b2bbaa421296301421ae8eb355 |
---|---|
record_format |
Article |
spelling |
doaj-a1d513b2bbaa421296301421ae8eb3552021-03-30T02:06:11ZengIEEEIEEE Access2169-35362020-01-018283492835910.1109/ACCESS.2020.29720558985332Joint Probabilistic People Detection in Overlapping Depth ImagesJohannes Wetzel0https://orcid.org/0000-0002-6869-9810Astrid Laubenheimer1https://orcid.org/0000-0001-7955-4521Michael Heizmann2https://orcid.org/0000-0001-9339-2055Intelligent Systems Research Group (ISRG), Karlsruhe University of Applied Sciences, Karlsruhe, GermanyIntelligent Systems Research Group (ISRG), Karlsruhe University of Applied Sciences, Karlsruhe, GermanyInstitute of Industrial Information Technology (IIIT), Karlsruhe Institute of Technology (KIT), Karlsruhe, GermanyPrivacy-preserving high-quality people detection is a vital computer vision task for various indoor scenarios, e.g. people counting, customer behavior analysis, ambient assisted living or smart homes. In this work a novel approach for people detection in multiple overlapping depth images is proposed. We present a probabilistic framework utilizing a generative scene model to jointly exploit the multi-view image evidence, allowing us to detect people from arbitrary viewpoints. Our approach makes use of mean-field variational inference to not only estimate the maximum a posteriori (MAP) state but to also approximate the posterior probability distribution of people present in the scene. Evaluation shows state-of-the-art results on a novel data set for indoor people detection and tracking in depth images from the top-view with high perspective distortions. Furthermore it can be demonstrated that our approach (compared to the the mono-view setup) successfully exploits the multi-view image evidence and robustly converges in only a few iterations.https://ieeexplore.ieee.org/document/8985332/Depth sensor indoor surveillancedepth sensor networksgenerative scene modeljoint multi-view person detectionmean-field variational inferencemulti-camera person detection |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Johannes Wetzel Astrid Laubenheimer Michael Heizmann |
spellingShingle |
Johannes Wetzel Astrid Laubenheimer Michael Heizmann Joint Probabilistic People Detection in Overlapping Depth Images IEEE Access Depth sensor indoor surveillance depth sensor networks generative scene model joint multi-view person detection mean-field variational inference multi-camera person detection |
author_facet |
Johannes Wetzel Astrid Laubenheimer Michael Heizmann |
author_sort |
Johannes Wetzel |
title |
Joint Probabilistic People Detection in Overlapping Depth Images |
title_short |
Joint Probabilistic People Detection in Overlapping Depth Images |
title_full |
Joint Probabilistic People Detection in Overlapping Depth Images |
title_fullStr |
Joint Probabilistic People Detection in Overlapping Depth Images |
title_full_unstemmed |
Joint Probabilistic People Detection in Overlapping Depth Images |
title_sort |
joint probabilistic people detection in overlapping depth images |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2020-01-01 |
description |
Privacy-preserving high-quality people detection is a vital computer vision task for various indoor scenarios, e.g. people counting, customer behavior analysis, ambient assisted living or smart homes. In this work a novel approach for people detection in multiple overlapping depth images is proposed. We present a probabilistic framework utilizing a generative scene model to jointly exploit the multi-view image evidence, allowing us to detect people from arbitrary viewpoints. Our approach makes use of mean-field variational inference to not only estimate the maximum a posteriori (MAP) state but to also approximate the posterior probability distribution of people present in the scene. Evaluation shows state-of-the-art results on a novel data set for indoor people detection and tracking in depth images from the top-view with high perspective distortions. Furthermore it can be demonstrated that our approach (compared to the the mono-view setup) successfully exploits the multi-view image evidence and robustly converges in only a few iterations. |
topic |
Depth sensor indoor surveillance depth sensor networks generative scene model joint multi-view person detection mean-field variational inference multi-camera person detection |
url |
https://ieeexplore.ieee.org/document/8985332/ |
work_keys_str_mv |
AT johanneswetzel jointprobabilisticpeopledetectioninoverlappingdepthimages AT astridlaubenheimer jointprobabilisticpeopledetectioninoverlappingdepthimages AT michaelheizmann jointprobabilisticpeopledetectioninoverlappingdepthimages |
_version_ |
1724185765007589376 |