Accuracy Evaluation of 3D Pose Reconstruction Algorithms Through Stereo Camera Information Fusion for Physical Exercises with MediaPipe Pose

In recent years, significant research has been conducted on video-based human pose estimation (HPE). While monocular two-dimensional (2D) HPE has been shown to achieve high performance, monocular three-dimensional (3D) HPE poses a more challenging problem. However, since human motion happens in a 3D...

Full description

Saved in:
Bibliographic Details
Main Authors: Sebastian Dill, Arjang Ahmadi, Martin Grimmer, Dennis Haufe, Maurice Rohr, Yanhua Zhao, Maziar Sharbafi, Christoph Hoog Antink
Format: Article
Language:English
Published: MDPI AG 2024-12-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/24/23/7772
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850106522220101632
author Sebastian Dill
Arjang Ahmadi
Martin Grimmer
Dennis Haufe
Maurice Rohr
Yanhua Zhao
Maziar Sharbafi
Christoph Hoog Antink
author_facet Sebastian Dill
Arjang Ahmadi
Martin Grimmer
Dennis Haufe
Maurice Rohr
Yanhua Zhao
Maziar Sharbafi
Christoph Hoog Antink
author_sort Sebastian Dill
collection DOAJ
description In recent years, significant research has been conducted on video-based human pose estimation (HPE). While monocular two-dimensional (2D) HPE has been shown to achieve high performance, monocular three-dimensional (3D) HPE poses a more challenging problem. However, since human motion happens in a 3D space, 3D HPE offers a more accurate representation of the human, granting increased usability for complex tasks like analysis of physical exercise. We propose a method based on MediaPipe Pose, 2D HPE on stereo cameras and a fusion algorithm without prior stereo calibration to reconstruct 3D poses, combining the advantages of high accuracy in 2D HPE with the increased usability of 3D coordinates. We evaluate this method on a self-recorded database focused on physical exercise to research what accuracy can be achieved and whether this accuracy is sufficient to recognize errors in exercise performance. We find that our method achieves significantly improved performance compared to monocular 3D HPE (median RMSE of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>30.1</mn></mrow></semantics></math></inline-formula> compared to <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>56.3</mn></mrow></semantics></math></inline-formula>, <i>p</i>-value below <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mn>10</mn><mrow><mo>−</mo><mn>6</mn></mrow></msup></semantics></math></inline-formula>) and can show that the performance is sufficient for error recognition.
format Article
id doaj-art-ca97c48fe6cb4ca5be611bb639058a83
institution OA Journals
issn 1424-8220
language English
publishDate 2024-12-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-ca97c48fe6cb4ca5be611bb639058a832025-08-20T02:38:48ZengMDPI AGSensors1424-82202024-12-012423777210.3390/s24237772Accuracy Evaluation of 3D Pose Reconstruction Algorithms Through Stereo Camera Information Fusion for Physical Exercises with MediaPipe PoseSebastian Dill0Arjang Ahmadi1Martin Grimmer2Dennis Haufe3Maurice Rohr4Yanhua Zhao5Maziar Sharbafi6Christoph Hoog Antink7KIS*MED (AI Systems in Medicine), Technische Universität Darmstadt, 64283 Darmstadt, GermanyLauflabor (Locomotion Laboratory), Centre for Cognitive Science, Technische Universität Darmstadt, 64289 Darmstadt, GermanyLauflabor (Locomotion Laboratory), Centre for Cognitive Science, Technische Universität Darmstadt, 64289 Darmstadt, GermanyLauflabor (Locomotion Laboratory), Centre for Cognitive Science, Technische Universität Darmstadt, 64289 Darmstadt, GermanyKIS*MED (AI Systems in Medicine), Technische Universität Darmstadt, 64283 Darmstadt, GermanyKIS*MED (AI Systems in Medicine), Technische Universität Darmstadt, 64283 Darmstadt, GermanyLauflabor (Locomotion Laboratory), Centre for Cognitive Science, Technische Universität Darmstadt, 64289 Darmstadt, GermanyKIS*MED (AI Systems in Medicine), Technische Universität Darmstadt, 64283 Darmstadt, GermanyIn recent years, significant research has been conducted on video-based human pose estimation (HPE). While monocular two-dimensional (2D) HPE has been shown to achieve high performance, monocular three-dimensional (3D) HPE poses a more challenging problem. However, since human motion happens in a 3D space, 3D HPE offers a more accurate representation of the human, granting increased usability for complex tasks like analysis of physical exercise. We propose a method based on MediaPipe Pose, 2D HPE on stereo cameras and a fusion algorithm without prior stereo calibration to reconstruct 3D poses, combining the advantages of high accuracy in 2D HPE with the increased usability of 3D coordinates. We evaluate this method on a self-recorded database focused on physical exercise to research what accuracy can be achieved and whether this accuracy is sufficient to recognize errors in exercise performance. We find that our method achieves significantly improved performance compared to monocular 3D HPE (median RMSE of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>30.1</mn></mrow></semantics></math></inline-formula> compared to <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>56.3</mn></mrow></semantics></math></inline-formula>, <i>p</i>-value below <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mn>10</mn><mrow><mo>−</mo><mn>6</mn></mrow></msup></semantics></math></inline-formula>) and can show that the performance is sufficient for error recognition.https://www.mdpi.com/1424-8220/24/23/7772computer visionhuman pose estimationinformation fusionMediaPipe Pose
spellingShingle Sebastian Dill
Arjang Ahmadi
Martin Grimmer
Dennis Haufe
Maurice Rohr
Yanhua Zhao
Maziar Sharbafi
Christoph Hoog Antink
Accuracy Evaluation of 3D Pose Reconstruction Algorithms Through Stereo Camera Information Fusion for Physical Exercises with MediaPipe Pose
Sensors
computer vision
human pose estimation
information fusion
MediaPipe Pose
title Accuracy Evaluation of 3D Pose Reconstruction Algorithms Through Stereo Camera Information Fusion for Physical Exercises with MediaPipe Pose
title_full Accuracy Evaluation of 3D Pose Reconstruction Algorithms Through Stereo Camera Information Fusion for Physical Exercises with MediaPipe Pose
title_fullStr Accuracy Evaluation of 3D Pose Reconstruction Algorithms Through Stereo Camera Information Fusion for Physical Exercises with MediaPipe Pose
title_full_unstemmed Accuracy Evaluation of 3D Pose Reconstruction Algorithms Through Stereo Camera Information Fusion for Physical Exercises with MediaPipe Pose
title_short Accuracy Evaluation of 3D Pose Reconstruction Algorithms Through Stereo Camera Information Fusion for Physical Exercises with MediaPipe Pose
title_sort accuracy evaluation of 3d pose reconstruction algorithms through stereo camera information fusion for physical exercises with mediapipe pose
topic computer vision
human pose estimation
information fusion
MediaPipe Pose
url https://www.mdpi.com/1424-8220/24/23/7772
work_keys_str_mv AT sebastiandill accuracyevaluationof3dposereconstructionalgorithmsthroughstereocamerainformationfusionforphysicalexerciseswithmediapipepose
AT arjangahmadi accuracyevaluationof3dposereconstructionalgorithmsthroughstereocamerainformationfusionforphysicalexerciseswithmediapipepose
AT martingrimmer accuracyevaluationof3dposereconstructionalgorithmsthroughstereocamerainformationfusionforphysicalexerciseswithmediapipepose
AT dennishaufe accuracyevaluationof3dposereconstructionalgorithmsthroughstereocamerainformationfusionforphysicalexerciseswithmediapipepose
AT mauricerohr accuracyevaluationof3dposereconstructionalgorithmsthroughstereocamerainformationfusionforphysicalexerciseswithmediapipepose
AT yanhuazhao accuracyevaluationof3dposereconstructionalgorithmsthroughstereocamerainformationfusionforphysicalexerciseswithmediapipepose
AT maziarsharbafi accuracyevaluationof3dposereconstructionalgorithmsthroughstereocamerainformationfusionforphysicalexerciseswithmediapipepose
AT christophhoogantink accuracyevaluationof3dposereconstructionalgorithmsthroughstereocamerainformationfusionforphysicalexerciseswithmediapipepose