Cargando…

Effect of Face Blurring on Human Pose Estimation: Ensuring Subject Privacy for Medical and Occupational Health Applications

The face blurring of images plays a key role in protecting privacy. However, in computer vision, especially for the human pose estimation task, machine-learning models are currently trained, validated, and tested on original datasets without face blurring. Additionally, the accuracy of human pose es...

Descripción completa

Detalles Bibliográficos
Autores principales: Jiang, Jindong, Skalli, Wafa, Siadat, Ali, Gajny, Laurent
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9739378/
https://www.ncbi.nlm.nih.gov/pubmed/36502076
http://dx.doi.org/10.3390/s22239376
_version_ 1784847790531149824
author Jiang, Jindong
Skalli, Wafa
Siadat, Ali
Gajny, Laurent
author_facet Jiang, Jindong
Skalli, Wafa
Siadat, Ali
Gajny, Laurent
author_sort Jiang, Jindong
collection PubMed
description The face blurring of images plays a key role in protecting privacy. However, in computer vision, especially for the human pose estimation task, machine-learning models are currently trained, validated, and tested on original datasets without face blurring. Additionally, the accuracy of human pose estimation is of great importance for kinematic analysis. This analysis is relevant in areas such as occupational safety and clinical gait analysis where privacy is crucial. Therefore, in this study, we explore the impact of face blurring on human pose estimation and the subsequent kinematic analysis. Firstly, we blurred the subjects’ heads in the image dataset. Then we trained our neural networks using the face-blurred and the original unblurred dataset. Subsequently, the performances of the different models, in terms of landmark localization and joint angles, were estimated on blurred and unblurred testing data. Finally, we examined the statistical significance of the effect of face blurring on the kinematic analysis along with the strength of the effect. Our results reveal that the strength of the effect of face blurring was low and within acceptable limits (<1°). We have thus shown that for human pose estimation, face blurring guarantees subject privacy while not degrading the prediction performance of a deep learning model.
format Online
Article
Text
id pubmed-9739378
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-97393782022-12-11 Effect of Face Blurring on Human Pose Estimation: Ensuring Subject Privacy for Medical and Occupational Health Applications Jiang, Jindong Skalli, Wafa Siadat, Ali Gajny, Laurent Sensors (Basel) Article The face blurring of images plays a key role in protecting privacy. However, in computer vision, especially for the human pose estimation task, machine-learning models are currently trained, validated, and tested on original datasets without face blurring. Additionally, the accuracy of human pose estimation is of great importance for kinematic analysis. This analysis is relevant in areas such as occupational safety and clinical gait analysis where privacy is crucial. Therefore, in this study, we explore the impact of face blurring on human pose estimation and the subsequent kinematic analysis. Firstly, we blurred the subjects’ heads in the image dataset. Then we trained our neural networks using the face-blurred and the original unblurred dataset. Subsequently, the performances of the different models, in terms of landmark localization and joint angles, were estimated on blurred and unblurred testing data. Finally, we examined the statistical significance of the effect of face blurring on the kinematic analysis along with the strength of the effect. Our results reveal that the strength of the effect of face blurring was low and within acceptable limits (<1°). We have thus shown that for human pose estimation, face blurring guarantees subject privacy while not degrading the prediction performance of a deep learning model. MDPI 2022-12-01 /pmc/articles/PMC9739378/ /pubmed/36502076 http://dx.doi.org/10.3390/s22239376 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/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 (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Jiang, Jindong
Skalli, Wafa
Siadat, Ali
Gajny, Laurent
Effect of Face Blurring on Human Pose Estimation: Ensuring Subject Privacy for Medical and Occupational Health Applications
title Effect of Face Blurring on Human Pose Estimation: Ensuring Subject Privacy for Medical and Occupational Health Applications
title_full Effect of Face Blurring on Human Pose Estimation: Ensuring Subject Privacy for Medical and Occupational Health Applications
title_fullStr Effect of Face Blurring on Human Pose Estimation: Ensuring Subject Privacy for Medical and Occupational Health Applications
title_full_unstemmed Effect of Face Blurring on Human Pose Estimation: Ensuring Subject Privacy for Medical and Occupational Health Applications
title_short Effect of Face Blurring on Human Pose Estimation: Ensuring Subject Privacy for Medical and Occupational Health Applications
title_sort effect of face blurring on human pose estimation: ensuring subject privacy for medical and occupational health applications
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9739378/
https://www.ncbi.nlm.nih.gov/pubmed/36502076
http://dx.doi.org/10.3390/s22239376
work_keys_str_mv AT jiangjindong effectoffaceblurringonhumanposeestimationensuringsubjectprivacyformedicalandoccupationalhealthapplications
AT skalliwafa effectoffaceblurringonhumanposeestimationensuringsubjectprivacyformedicalandoccupationalhealthapplications
AT siadatali effectoffaceblurringonhumanposeestimationensuringsubjectprivacyformedicalandoccupationalhealthapplications
AT gajnylaurent effectoffaceblurringonhumanposeestimationensuringsubjectprivacyformedicalandoccupationalhealthapplications