Camera Setup and OpenPose Software without GPU for Calibration and Recording in Telerehabilitation Use

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21460%2F21%3A00354734" target="_blank" >RIV/68407700:21460/21:00354734 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/68407700:21730/21:00354734

Výsledek na webu

<a href="https://doi.org/10.1109/EHB52898.2021.9657743" target="_blank" >https://doi.org/10.1109/EHB52898.2021.9657743</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/EHB52898.2021.9657743" target="_blank" >10.1109/EHB52898.2021.9657743</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Camera Setup and OpenPose Software without GPU for Calibration and Recording in Telerehabilitation Use

Popis výsledku v původním jazyce

Home exercises are significant in the rehabilitation process of physiotherapy patients, which lack immediate feedback as to the proper movement and therefore might humper patient treatment. In this paper we are proposing an algorithm for fast tracking of human body movements performed during physiotherapeutic exercises using a simple webcam home setup and common domestically available CPU computing resources. We use OpenPose for detecting body vertices in key frames and a novel vertex tracking algorithm between video frames, which leverages encoded video Motion Vectors (MVs). We show excellent tracking accuracy between frames and x15 reduction in time, as compared to native OpenPose, which would require a Graphical Processing Unit (GPU) to perform in real time. We further provide a design and implementation of a precision camera system consisting of two cameras in the frontal and lateral direction, which were precisely positioned using a laser cross. This system will be also used to verify whether the webcam is able to record with sufficient quality to further image processing analysis. As part of this work, a camera system including supporting calibration and recording scripts was designed and implemented. The cameras triggers were synchronized by wire interconnection and set up by proposed script. In this work we synchronize two cameras and align their frames such that the OpenPose can be applied independently to each of the two channels to measure movements from two different body projections (3D).

Název v anglickém jazyce

Camera Setup and OpenPose Software without GPU for Calibration and Recording in Telerehabilitation Use

Popis výsledku anglicky

Home exercises are significant in the rehabilitation process of physiotherapy patients, which lack immediate feedback as to the proper movement and therefore might humper patient treatment. In this paper we are proposing an algorithm for fast tracking of human body movements performed during physiotherapeutic exercises using a simple webcam home setup and common domestically available CPU computing resources. We use OpenPose for detecting body vertices in key frames and a novel vertex tracking algorithm between video frames, which leverages encoded video Motion Vectors (MVs). We show excellent tracking accuracy between frames and x15 reduction in time, as compared to native OpenPose, which would require a Graphical Processing Unit (GPU) to perform in real time. We further provide a design and implementation of a precision camera system consisting of two cameras in the frontal and lateral direction, which were precisely positioned using a laser cross. This system will be also used to verify whether the webcam is able to record with sufficient quality to further image processing analysis. As part of this work, a camera system including supporting calibration and recording scripts was designed and implemented. The cameras triggers were synchronized by wire interconnection and set up by proposed script. In this work we synchronize two cameras and align their frames such that the OpenPose can be applied independently to each of the two channels to measure movements from two different body projections (3D).

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20601 - Medical engineering

Projekt

<a href="/cs/project/LTAIZ19008" target="_blank" >LTAIZ19008: Zkvalitnění robotické fyzioterapeutické léčby pomocí metod strojového učení</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2021

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

2021 International Conference on e-Health and Bioengineering (EHB)

ISBN

978-1-6654-4000-4

ISSN

2575-5137

e-ISSN

2575-5145

Počet stran výsledku

4

Strana od-do

1-4

Název nakladatele

IEEE Industrial Electronic Society

Místo vydání

Vienna

Místo konání akce

Iasi

Datum konání akce

18. 11. 2021

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000802227900203