{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,14]],"date-time":"2026-11-14T02:54:10Z","timestamp":1794624850926,"version":"4.9.1"},"reference-count":77,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,2,7]],"date-time":"2025-02-07T00:00:00Z","timestamp":1738886400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,2,7]],"date-time":"2025-02-07T00:00:00Z","timestamp":1738886400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"EMHSeed program at the University of Toronto"},{"name":"Center for Advancing Neurotechnological Innovation to Application (CRANIA) at the University Health Network."}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BioMed Eng OnLine"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background and objectives<\/jats:title>\n                    <jats:p>Markerless vision-based human pose estimation (HPE) is a promising avenue towards scalable data collection in rehabilitation. Deploying this technology will require self-contained systems able to process data efficiently and accurately. The aims of this work are to (1) Determine how depth data affects lightweight monocular red\u2013green\u2013blue (RGB) HPE performance (accuracy and speed), to inform sensor selection and (2) Validate HPE models using data from individuals with physical impairments.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Two HPE models were investigated: Dite-HRNet and MobileHumanPose (capable of 2D and 3D HPE, respectively). The models were modified to include depth data as an input using three different fusion techniques: an early fusion method, a simple intermediate fusion method (using concatenation), and a complex intermediate fusion method (using specific fusion blocks, additional convolutional layers, and concatenation). All fusion techniques used RGB-D data, in contrast to the original models which only used RGB data. The models were trained, validated and tested using the CMU Panoptic and Human3.6\u00a0M data sets as well as a custom data set. The custom data set includes RGB-D and optical motion capture data of 15 uninjured and 12 post-stroke individuals, while they performed movements involving their upper limbs. HPE model performances were monitored through accuracy and computational efficiency. Evaluation metrics include Mean per Joint Position Error (MPJPE), Floating Point Operations (FLOPs) and frame rates (frames per second).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      The early fusion architecture consistently delivered the lowest MPJPE in both 2D and 3D HPE cases while achieving similar FLOPs and frame rates to its RGB counterpart. These results were consistent regardless of the data used for training and testing the HPE models. Comparisons between the uninjured and stroke groups did not reveal a significant effect (all\n                      <jats:italic>p<\/jats:italic>\n                      values\u2009&gt;\u20090.36) of motor impairment on the accuracy of any model.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Including depth data using an early fusion architecture improves the accuracy\u2013efficiency trade-off of the HPE model. HPE accuracy is not affected by the presence of physical impairments. These results suggest that using depth data with RGB data is beneficial to HPE, and that models trained with data collected from uninjured individuals can generalize to persons with physical impairments.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1186\/s12938-025-01347-y","type":"journal-article","created":{"date-parts":[[2025,2,7]],"date-time":"2025-02-07T02:16:01Z","timestamp":1738894561000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":13,"title":["The effect of depth data and upper limb impairment on lightweight monocular RGB human pose estimation models"],"prefix":"10.1186","volume":"24","author":[{"given":"Gloria-Edith","family":"Boudreault-Morales","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Cesar","family":"Marquez-Chin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xilin","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jos\u00e9","family":"Zariffa","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,2,7]]},"reference":[{"issue":"6","key":"1347_CR1","doi-asserted-by":"publisher","first-page":"828","DOI":"10.1097\/WCO.0000000000000750","volume":"32","author":"A-S Hofer","year":"2019","unstructured":"Hofer A-S, Schwab ME. Enhancing rehabilitation and functional recovery after brain and spinal cord trauma with electrical neuromodulation. Curr Opin Neurol. 2019;32(6):828\u201335.","journal-title":"Curr Opin Neurol"},{"key":"1347_CR2","doi-asserted-by":"publisher","DOI":"10.3390\/s20185104","author":"JA Albert","year":"2020","unstructured":"Albert JA, Owolabi V, Gebel A, Brahms CM, Granacher U, Arnrich B. Evaluation of the pose tracking performance of the azure Kinect and Kinect v2 for Gait analysis in comparison with a gold standard: a pilot study. Sensors. 2020. https:\/\/doi.org\/10.3390\/s20185104.","journal-title":"Sensors"},{"key":"1347_CR3","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2203.08792","author":"R James Cotton","year":"2022","unstructured":"James Cotton R. PosePipe: open-source human pose estimation pipeline for clinical research. arXiv. 2022. https:\/\/doi.org\/10.48550\/arXiv.2203.08792.","journal-title":"arXiv"},{"key":"1347_CR4","doi-asserted-by":"publisher","DOI":"10.1016\/j.displa.2022.102225","volume":"74","author":"D Pascual-Hern\u00e1ndez","year":"2022","unstructured":"Pascual-Hern\u00e1ndez D, Oyaga de Frutos N, Mora-Jim\u00e9nez I, Ca\u00f1as-Plaza JM. Efficient 3D human pose estimation from RGBD sensors. Displays. 2022;74: 102225.","journal-title":"Displays"},{"issue":"2","key":"1347_CR5","doi-asserted-by":"publisher","first-page":"3362","DOI":"10.3390\/s140203362","volume":"14","author":"A Muro-de-la-Herran","year":"2014","unstructured":"Muro-de-la-Herran A, Garcia-Zapirain B, Mendez-Zorrilla A. Gait analysis methods: an overview of wearable and non-wearable systems, highlighting clinical applications. Sensors (Basel). 2014;14(2):3362\u201394.","journal-title":"Sensors (Basel)"},{"key":"1347_CR6","doi-asserted-by":"crossref","unstructured":"Y Li, C Wang, Y Cao, B Liu, J Tan, Y Luo. \u201cHuman pose estimation based in-home lower body rehabilitation system.\u201d in 2020 International Joint Conference on Neural Networks (IJCNN). 2020;1\u20138.","DOI":"10.1109\/IJCNN48605.2020.9207296"},{"key":"1347_CR7","doi-asserted-by":"publisher","DOI":"10.1016\/j.cviu.2021.103275","volume":"212","author":"Y Desmarais","year":"2021","unstructured":"Desmarais Y, Mottet D, Slangen P, Montesinos P. A review of 3D human pose estimation algorithms for markerless motion capture. Comput Vis Image Underst. 2021;212: 103275.","journal-title":"Comput Vis Image Underst"},{"key":"1347_CR8","doi-asserted-by":"crossref","unstructured":"S Mroz et al. \u201cComparing the Quality of Human Pose Estimation with BlazePose or OpenPose.\u201d in 2021 4th International Conference on Bio-Engineering for Smart Technologies (BioSMART). 2021;1\u20134.","DOI":"10.1109\/BioSMART54244.2021.9677850"},{"key":"1347_CR9","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1804.06208","author":"B Xiao","year":"2018","unstructured":"Xiao B, Wu H, Wei Y. Simple baselines for human pose estimation and tracking. arXiv. 2018. https:\/\/doi.org\/10.48550\/arXiv.1804.06208.","journal-title":"arXiv"},{"issue":"4","key":"1347_CR10","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3524497","volume":"55","author":"W Liu","year":"2022","unstructured":"Liu W, Bao Q, Sun Y, Mei T. Recent advances of monocular 2D and 3D human pose estimation: a deep learning perspective. ACM Comput Surv. 2022;55(4):1\u201341.","journal-title":"ACM Comput Surv"},{"key":"1347_CR11","first-page":"380","volume-title":"Artificial Neural Networks and Machine Learning","author":"BC Dos Santos Mel\u00edcio","year":"2021","unstructured":"Dos Santos Mel\u00edcio BC, Baranyi G, Ga\u00e1l Z, Zidan S, L\u0151rincz A. DeepRehab real time pose estimation on the edge for knee injury rehabilitation. In: Artificial Neural Networks and Machine Learning. Cham: Springer International Publishing; 2021. p. 380\u201391."},{"key":"1347_CR12","doi-asserted-by":"publisher","DOI":"10.3390\/s21217315","author":"J Stenum","year":"2021","unstructured":"Stenum J, Cherry-Allen KM, Pyles CO, Reetzke RD, Vignos MF, Roemmich RT. Applications of Pose Estimation in Human Health and Performance across the Lifespan. Sensors. 2021. https:\/\/doi.org\/10.3390\/s21217315.","journal-title":"Sensors"},{"issue":"1","key":"1347_CR13","doi-asserted-by":"publisher","first-page":"97","DOI":"10.1186\/s12984-018-0446-z","volume":"15","author":"MH Li","year":"2018","unstructured":"Li MH, Mestre TA, Fox SH, Taati B. Vision-based assessment of parkinsonism and levodopa-induced dyskinesia with pose estimation. J Neuroeng Rehabil. 2018;15(1):97.","journal-title":"J Neuroeng Rehabil"},{"issue":"8","key":"1347_CR14","doi-asserted-by":"publisher","first-page":"29","DOI":"10.1007\/s11910-020-01049-z","volume":"20","author":"A Arac","year":"2020","unstructured":"Arac A. Machine learning for 3D kinematic analysis of movements in neurorehabilitation. Curr Neurol Neurosci Rep. 2020;20(8):29.","journal-title":"Curr Neurol Neurosci Rep"},{"issue":"1","key":"1347_CR15","doi-asserted-by":"publisher","first-page":"13","DOI":"10.2340\/1650197771331","volume":"7","author":"AR Fugl-Meyer","year":"1975","unstructured":"Fugl-Meyer AR, J\u00e4\u00e4sk\u00f6 L, Leyman I, Olsson S, Steglind S. The post-stroke hemiplegic patient. 1. A method for evaluation of physical performance. Scand J Rehabil Med. 1975;7(1):13\u201331.","journal-title":"Scand J Rehabil Med"},{"issue":"4","key":"1347_CR16","doi-asserted-by":"publisher","first-page":"483","DOI":"10.1097\/00004356-198112000-00001","volume":"4","author":"RC Lyle","year":"1981","unstructured":"Lyle RC. A performance test for assessment of upper limb function in physical rehabilitation treatment and research. Int J Rehabil Res. 1981;4(4):483\u201392.","journal-title":"Int J Rehabil Res"},{"key":"1347_CR17","doi-asserted-by":"publisher","first-page":"20","DOI":"10.3389\/fnsys.2019.00020","volume":"13","author":"A Arac","year":"2019","unstructured":"Arac A, Zhao P, Dobkin BH, Carmichael ST, Golshani P. DeepBehavior: a deep learning toolbox for automated analysis of animal and human behavior imaging data. Front Syst Neurosci. 2019;13:20.","journal-title":"Front Syst Neurosci"},{"issue":"1","key":"1347_CR18","doi-asserted-by":"publisher","first-page":"20673","DOI":"10.1038\/s41598-021-00212-x","volume":"11","author":"L Needham","year":"2021","unstructured":"Needham L, et al. The accuracy of several pose estimation methods for 3D joint centre localisation. Sci Rep. 2021;11(1):20673.","journal-title":"Sci Rep"},{"issue":"9","key":"1347_CR19","doi-asserted-by":"publisher","first-page":"1281","DOI":"10.1038\/s41593-018-0209-y","volume":"21","author":"A Mathis","year":"2018","unstructured":"Mathis A, et al. DeepLabCut: markerless pose estimation of user-defined body parts with deep learning. Nat Neurosci. 2018;21(9):1281\u20139.","journal-title":"Nat Neurosci"},{"key":"1347_CR20","first-page":"115","volume-title":"Proceedings of the 10th International Conference on PErvasive Technologies Related to Assistive Environments, in PETRA \u201917","author":"D Michel","year":"2017","unstructured":"Michel D, Qammaz A, Argyros AA. Markerless 3D Human Pose Estimation and Tracking based on RGBD Cameras: an Experimental Evaluation. In: Proceedings of the 10th International Conference on PErvasive Technologies Related to Assistive Environments, in PETRA \u201917. New York: Association for Computing Machinery; 2017. p. 115\u201322."},{"key":"1347_CR21","unstructured":"\u201cHuman pose estimation for Kinect,\u201d Microsoft Research. 2011. https:\/\/www.microsoft.com\/en-us\/research\/project\/human-pose-estimation-for-kinect\/. Accessed 04 Sep 2024."},{"issue":"1","key":"1347_CR22","doi-asserted-by":"publisher","first-page":"39","DOI":"10.1016\/j.jvcir.2013.03.011","volume":"25","author":"K Buys","year":"2014","unstructured":"Buys K, Cagniart C, Baksheev A, De Laet T, De Schutter J, Pantofaru C. An adaptable system for RGB-D based human body detection and pose estimation. J Vis Commun Image Represent. 2014;25(1):39\u201352.","journal-title":"J Vis Commun Image Represent"},{"key":"1347_CR23","doi-asserted-by":"publisher","unstructured":"A Jalal, Y Kim, D Kim. \u201cRidge body parts features for human pose estimation and recognition from RGB-D video data,\u201d in Fifth International Conference on Computing, Communications and Networking Technologies (ICCCNT). IEEE. 2014. https:\/\/doi.org\/10.1109\/icccnt.2014.6963015.","DOI":"10.1109\/icccnt.2014.6963015"},{"key":"1347_CR24","doi-asserted-by":"crossref","unstructured":"R Bashirov et al. \u201cReal-time rgbd-based extended body pose estimation.\u201d in Proceedings of the IEEE\/CVF Winter Conference on Applications of Computer Vision. 2021;2807\u20132816.","DOI":"10.1109\/WACV48630.2021.00285"},{"key":"1347_CR25","doi-asserted-by":"crossref","unstructured":"C Zimmermann, T Welschehold, C Dornhege, W Burgard, T Brox. \u201c3D human pose estimation in RGBD images for robotic task learning.\u201d in 2018 IEEE International Conference on Robotics and Automation (ICRA). 2018;1986\u20131992.","DOI":"10.1109\/ICRA.2018.8462833"},{"key":"1347_CR26","first-page":"5802","volume":"2020","author":"Q Wu","year":"2020","unstructured":"Wu Q, Xu G, Zhang S, Li Y, Wei F. \u201cHuman 3D pose estimation in a lying position by RGB-D images for medical diagnosis and rehabilitation.\u201d Conf Proc IEEE Eng Med Biol Soc. 2020;2020:5802\u20135.","journal-title":"Conf Proc IEEE Eng Med Biol Soc"},{"issue":"6","key":"1347_CR27","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3533384","volume":"55","author":"LK Topham","year":"2023","unstructured":"Topham LK, Khan W, Al-Jumeily D, Hussain A. Human body pose estimation for gait identification: a comprehensive survey of datasets and models. ACM Comput Surv. 2023;55(6):1\u201342.","journal-title":"ACM Comput Surv"},{"key":"1347_CR28","doi-asserted-by":"publisher","DOI":"10.1016\/j.cviu.2021.103225","volume":"210","author":"J Wang","year":"2021","unstructured":"Wang J, et al. Deep 3D human pose estimation: a review. Comput Vis Image Underst. 2021;210: 103225.","journal-title":"Comput Vis Image Underst"},{"issue":"3","key":"1347_CR29","doi-asserted-by":"publisher","first-page":"207","DOI":"10.1016\/j.vrih.2021.05.002","volume":"3","author":"L Huang","year":"2021","unstructured":"Huang L, Zhang B, Guo Z, Xiao Y, Cao Z, Yuan J. Survey on depth and RGB image-based 3D hand shape and pose estimation. Virtual Reality Intell Hardware. 2021;3(3):207\u201334.","journal-title":"Virtual Reality Intell Hardware"},{"key":"1347_CR30","doi-asserted-by":"publisher","first-page":"258","DOI":"10.1007\/978-3-030-61609-0_21","volume-title":"Artificial Neural Networks and Machine Learning\u2014ICANN 2020","author":"M V\u00e9ges","year":"2020","unstructured":"V\u00e9ges M, L\u0151rincz A. Multi-person absolute 3D Human pose estimation with weak depth supervision. In: Artificial Neural Networks and Machine Learning\u2014ICANN 2020. Berlin: Springer International Publishing; 2020. p. 258\u201370."},{"key":"1347_CR31","doi-asserted-by":"crossref","unstructured":"J Ying, X Zhao. \u201cRgb-D fusion for point-cloud-based 3d human pose estimation.\u201d in 2021 IEEE International Conference on Image Processing (ICIP). 2021; 3108\u20133112.","DOI":"10.1109\/ICIP42928.2021.9506588"},{"key":"1347_CR32","first-page":"740","volume-title":"Computer Vision\u2014ECCV","author":"T-Y Lin","year":"2014","unstructured":"Lin T-Y, et al. Microsoft COCO common objects in context. In: Computer Vision\u2014ECCV. Berlin: Springer International Publishing; 2014. p. 740\u201355."},{"key":"1347_CR33","first-page":"3686","volume-title":"IEEE Conference on Computer Vision and Pattern Recognition","author":"M Andriluka","year":"2014","unstructured":"Andriluka M, Pishchulin L, Gehler P, Schiele B. 2D human pose estimation: new benchmark and state of the art analysis. In: IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE; 2014. p. 3686\u201393."},{"issue":"7","key":"1347_CR34","doi-asserted-by":"publisher","first-page":"1325","DOI":"10.1109\/TPAMI.2013.248","volume":"36","author":"C Ionescu","year":"2014","unstructured":"Ionescu C, Papava D, Olaru V, Sminchisescu C. Human3.6M: large scale datasets and predictive methods for 3D human sensing in natural environments. IEEE Trans Pattern Anal Mach Intell. 2014;36(7):1325\u201339.","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"issue":"1","key":"1347_CR35","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1007\/s11263-009-0273-6","volume":"87","author":"L Sigal","year":"2010","unstructured":"Sigal L, Balan AO, Black MJ. HumanEva: synchronized video and motion capture dataset and baseline algorithm for evaluation of articulated human motion. Int J Comput Vis. 2010;87(1):4\u201327.","journal-title":"Int J Comput Vis"},{"key":"1347_CR36","doi-asserted-by":"publisher","DOI":"10.3390\/s20215986","author":"I Doroniewicz","year":"2020","unstructured":"Doroniewicz I, et al. Writhing movement detection in newborns on the second and third day of life using pose-based feature machine learning classification. Sensors. 2020. https:\/\/doi.org\/10.3390\/s20215986.","journal-title":"Sensors"},{"issue":"1","key":"1347_CR37","doi-asserted-by":"publisher","first-page":"57","DOI":"10.1186\/s12984-023-01186-9","volume":"20","author":"WWT Lam","year":"2023","unstructured":"Lam WWT, Tang YM, Fong KNK. A systematic review of the applications of markerless motion capture (MMC) technology for clinical measurement in rehabilitation. J Neuroeng Rehabil. 2023;20(1):57.","journal-title":"J Neuroeng Rehabil"},{"issue":"2","key":"1347_CR38","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3328925","volume":"3","author":"R Langevin","year":"2019","unstructured":"Langevin R, et al. The PARK framework for automated analysis of parkinson\u2019s disease characteristics. Proc ACM Interact Mob Wearable Ubiquitous Technol. 2019;3(2):1\u201322.","journal-title":"Proc ACM Interact Mob Wearable Ubiquitous Technol"},{"key":"1347_CR39","doi-asserted-by":"publisher","DOI":"10.1016\/j.jneumeth.2019.108576","volume":"333","author":"Y Pang","year":"2020","unstructured":"Pang Y, et al. Automatic detection and quantification of hand movements toward development of an objective assessment of tremor and bradykinesia in Parkinson\u2019s disease. J Neurosci Methods. 2020;333: 108576.","journal-title":"J Neurosci Methods"},{"key":"1347_CR40","first-page":"5830364","volume":"2018","author":"HR Martinez","year":"2018","unstructured":"Martinez HR, Garcia-Sarreon A, Camara-Lemarroy C, Salazar F, Guerrero-Gonz\u00e1lez ML. Accuracy of markerless 3D motion capture evaluation to differentiate between On\/Off Status in Parkinson\u2019s Disease after deep brain stimulation. Parkinsons Dis. 2018;2018:5830364.","journal-title":"Parkinsons Dis"},{"issue":"3","key":"1347_CR41","doi-asserted-by":"publisher","first-page":"7033","DOI":"10.1016\/j.eswa.2008.08.076","volume":"36","author":"C-W Cho","year":"2009","unstructured":"Cho C-W, Chao W-H, Lin S-H, Chen Y-Y. A vision-based analysis system for gait recognition in patients with Parkinson\u2019s disease. Expert Syst Appl. 2009;36(3):7033\u20139.","journal-title":"Expert Syst Appl"},{"issue":"8","key":"1347_CR42","doi-asserted-by":"publisher","first-page":"773","DOI":"10.1111\/j.1469-8749.2010.03629.x","volume":"52","author":"L Adde","year":"2010","unstructured":"Adde L, Helbostad JL, Jensenius AR, Taraldsen G, Grunewaldt KH, St\u00f8en R. Early prediction of cerebral palsy by computer-based video analysis of general movements: a feasibility study. Dev Med Child Neurol. 2010;52(8):773\u20138.","journal-title":"Dev Med Child Neurol"},{"key":"1347_CR43","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1612.03153","author":"H Joo","year":"2016","unstructured":"Joo H, et al. Panoptic studio: a massively multiview system for social interaction capture. arXiv. 2016. https:\/\/doi.org\/10.48550\/arXiv.1612.03153.","journal-title":"arXiv"},{"key":"1347_CR44","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2204.10762","author":"Q Li","year":"2022","unstructured":"Li Q, Zhang Z, Xiao F, Zhang F, Bhanu B. Dite-HRNet dynamic lightweight high-resolution network for human pose estimation. arXiv. 2022. https:\/\/doi.org\/10.48550\/arXiv.2204.10762.","journal-title":"arXiv"},{"key":"1347_CR45","doi-asserted-by":"crossref","unstructured":"S Choi, S Choi, C Kim. \u201cMobileHumanPose: toward real-time 3D human pose estimation in mobile devices.\u201d in Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition. 2021; 2328\u20132338.","DOI":"10.1109\/CVPRW53098.2021.00265"},{"issue":"1","key":"1347_CR46","doi-asserted-by":"publisher","first-page":"172","DOI":"10.1109\/TPAMI.2019.2929257","volume":"43","author":"Z Cao","year":"2021","unstructured":"Cao Z, Hidalgo G, Simon T, Wei S-E, Sheikh Y. OpenPose: realtime multi-person 2D pose estimation using part affinity fields. IEEE Trans Pattern Anal Mach Intell. 2021;43(1):172\u201386.","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"1347_CR47","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2204.07370","author":"H Chen","year":"2022","unstructured":"Chen H, Feng R, Wu S, Xu H, Zhou F, Liu Z. 2D human pose estimation: a survey. arXiv. 2022. https:\/\/doi.org\/10.48550\/arXiv.2204.07370.","journal-title":"arXiv"},{"issue":"1","key":"1347_CR48","doi-asserted-by":"publisher","first-page":"167","DOI":"10.1007\/s00530-022-00980-0","volume":"29","author":"S Dubey","year":"2023","unstructured":"Dubey S, Dixit M. A comprehensive survey on human pose estimation approaches. Multimed Syst. 2023;29(1):167\u201395.","journal-title":"Multimed Syst"},{"key":"1347_CR49","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2406.19290","author":"P Knap","year":"2024","unstructured":"Knap P. Human modelling and pose estimation overview. arXiv. 2024. https:\/\/doi.org\/10.48550\/arXiv.2406.19290.","journal-title":"arXiv"},{"issue":"2","key":"1347_CR50","doi-asserted-by":"publisher","first-page":"123","DOI":"10.1080\/09540099550039318","volume":"7","author":"A Robins","year":"1995","unstructured":"Robins A. Catastrophic forgetting, rehearsal and pseudorehearsal. Conn Sci. 1995;7(2):123\u201346.","journal-title":"Conn Sci"},{"key":"1347_CR51","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1612.00796","author":"J Kirkpatrick","year":"2016","unstructured":"Kirkpatrick J, et al. Overcoming catastrophic forgetting in neural networks. arXiv. 2016. https:\/\/doi.org\/10.48550\/arXiv.1612.00796.","journal-title":"arXiv"},{"key":"1347_CR52","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2312.10549","author":"EL Aleixo","year":"2023","unstructured":"Aleixo EL, Colonna JG, Cristo M, Fernandes E. Catastrophic forgetting in deep learning: a comprehensive taxonomy. ArXiv. 2023. https:\/\/doi.org\/10.48550\/arXiv.2312.10549.","journal-title":"ArXiv"},{"issue":"10","key":"1347_CR53","first-page":"2265","volume":"12","author":"H-G Doan","year":"2023","unstructured":"Doan H-G, Luong H-Q, Ha T-O, Pham TTT. An efficient strategy for catastrophic forgetting reduction in incremental learning. Electronics (Basel). 2023;12(10):2265.","journal-title":"Electronics (Basel)"},{"key":"1347_CR54","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1904.00310","author":"X Li","year":"2019","unstructured":"Li X, Zhou Y, Wu T, Socher R, Xiong C. Learn to grow: a continual structure learning framework for overcoming catastrophic forgetting. arXiv. 2019. https:\/\/doi.org\/10.48550\/arXiv.1904.00310.","journal-title":"arXiv"},{"key":"1347_CR55","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2110.13598","author":"GK Nayak","year":"2021","unstructured":"Nayak GK, Shah H, Chakraborty A. Incremental learning for animal pose estimation using RBF k-DPP. arXiv. 2021. https:\/\/doi.org\/10.48550\/arXiv.2110.13598.","journal-title":"arXiv"},{"key":"1347_CR56","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2007.09389","author":"A Zeng","year":"2020","unstructured":"Zeng A, Sun X, Huang F, Liu M, Xu Q, Lin S. SRNet: improving generalization in 3D human pose estimation with a split-and-recombine approach. arXiv. 2020. https:\/\/doi.org\/10.48550\/arXiv.2007.09389.","journal-title":"arXiv"},{"key":"1347_CR57","doi-asserted-by":"publisher","unstructured":"C Li, GH. Lee. \u201cGenerating multiple hypotheses for 3D human pose estimation with mixture density network,\u201d in 2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE. 2019. https:\/\/doi.org\/10.1109\/cvpr.2019.01012.","DOI":"10.1109\/cvpr.2019.01012"},{"key":"1347_CR58","first-page":"1","volume":"2023","author":"RJ Cotton","year":"2023","unstructured":"Cotton RJ, et al. Optimizing trajectories and inverse kinematics for biomechanical analysis of markerless motion capture data. IEEE Int Conf Rehabil Robot. 2023;2023:1\u20136.","journal-title":"IEEE Int Conf Rehabil Robot"},{"key":"1347_CR59","unstructured":"README_kinoptic.Md at master \u00b7 CMU-Perceptual-Computing-Lab\/panoptic-toolbox. Github. https:\/\/github.com\/CMU-Perceptual-Computing-Lab\/panoptic-toolbox\/blob\/master\/README_kinoptic.md. Accessed 17 Jul 2024."},{"key":"1347_CR60","unstructured":"S Khanna. \u201cA comprehensive guide to train-test-validation split in 2024,\u201d Analytics Vidhya. 2023. https:\/\/www.analyticsvidhya.com\/blog\/2023\/11\/train-test-validation-split\/. Accessed 17 2024."},{"key":"1347_CR61","doi-asserted-by":"publisher","first-page":"104282","DOI":"10.1016\/j.imavis.2021.104282","volume":"114","author":"M Ben Gamra","year":"2021","unstructured":"Ben Gamra M, Akhloufi MA. A review of deep learning techniques for 2D and 3D human pose estimation. Image Vis Comput. 2021;114:104282.","journal-title":"Image Vis Comput"},{"key":"1347_CR62","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2207.02696","author":"C-Y Wang","year":"2022","unstructured":"Wang C-Y, Bochkovskiy A, Liao H-YM. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors. arXiv. 2022. https:\/\/doi.org\/10.48550\/arXiv.2207.02696.","journal-title":"arXiv"},{"key":"1347_CR63","unstructured":"\u201cOpenCV: RGB-Depth Processing.\u201d https:\/\/docs.opencv.org\/4.x\/d2\/d3a\/group__rgbd.html. Accessed 18 Jul 2024."},{"key":"1347_CR64","unstructured":"Z Zhang. Dite-HRNet: [IJCAI 2022] Code for the paper \u201cDite-HRNet: Dynamic Lightweight High-Resolution Network for Human Pose Estimation.\u201d Github. https:\/\/github.com\/ZiyiZhang27\/Dite-HRNet. Accessed 18 Jul 2024."},{"key":"1347_CR65","unstructured":"G. Pavlakos, C2f-vol-demo. Github. https:\/\/github.com\/geopavlakos\/c2f-vol-demo\/issues\/6. Accessed 18 Jul 2024."},{"key":"1347_CR66","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbab569","author":"SR Stahlschmidt","year":"2022","unstructured":"Stahlschmidt SR, Ulfenborg B, Synnergren J. Multimodal deep learning for biomedical data fusion: a review. Brief Bioinform. 2022. https:\/\/doi.org\/10.1093\/bib\/bbab569.","journal-title":"Brief Bioinform"},{"issue":"109165","key":"1347_CR67","doi-asserted-by":"publisher","DOI":"10.1016\/j.sigpro.2023.109165","volume":"213","author":"Q Tang","year":"2023","unstructured":"Tang Q, Liang J, Zhu F. A comparative review on multi-modal sensors fusion based on deep learning. Signal Process. 2023;213(109165): 109165.","journal-title":"Signal Process"},{"key":"1347_CR68","doi-asserted-by":"crossref","unstructured":"Z Wang, Y Lu, W Ni, L Song. \u201cAn RGB-D Based Approach for Human Pose Estimation.\u201d in 2021 International Conference on Networking Systems of AI (INSAI). 2021;166\u2013170.","DOI":"10.1109\/INSAI54028.2021.00039"},{"key":"1347_CR69","unstructured":"Tools\/analysis\/get_flops.Py at 0.X \u00b7 open-mmlab\/mmpose. Github. Available: https:\/\/github.com\/open-mmlab\/mmpose\/blob\/0.x\/tools\/analysis\/get_flops.py. Accessed 12 Aug 2024."},{"key":"1347_CR70","unstructured":"MrYXJ, calculate-flops.pytorch: The calflops is designed to calculate FLOPs\u3001MACs and Parameters in all various neural networks, such as Linear\u3001 CNN\u3001 RNN\u3001 GCN\u3001Transformer(Bert\u3001LlaMA etc Large Language Model). Github. https:\/\/github.com\/MrYxJ\/calculate-flops.pytorch. Accessed 12 Aug 2024."},{"issue":"11","key":"1347_CR71","doi-asserted-by":"publisher","first-page":"2493","DOI":"10.1161\/01.STR.0000185928.90848.2e","volume":"36","author":"G Uswatte","year":"2005","unstructured":"Uswatte G, Taub E, Morris D, Vignolo M, McCulloch K. Reliability and validity of the upper-extremity Motor Activity Log-14 for measuring real-world arm use. Stroke. 2005;36(11):2493\u20136.","journal-title":"Stroke"},{"key":"1347_CR72","doi-asserted-by":"crossref","unstructured":"AR Fugl-Meyer, L Jaasko, I Leyman, S Olsson, S Steglind. \u201cFugl-Meyer assessment upper extremity (FMA-UE) assessment of sensorimotor function.\u201d Rehabilitation Medicine, University of Gothenburg. 2019.","DOI":"10.4324\/9781003076391-112"},{"key":"1347_CR73","unstructured":"\u201cAward winning Motion Capture Systems,\u201d Vicon. 2019. https:\/\/www.vicon.com\/. Accessed 18 Jul 2024."},{"issue":"3","key":"1347_CR74","doi-asserted-by":"publisher","first-page":"300","DOI":"10.1016\/j.gaitpost.2009.11.007","volume":"31","author":"A Murgia","year":"2010","unstructured":"Murgia A, Kyberd P, Barnhill T. The use of kinematic and parametric information to highlight lack of movement and compensation in the upper extremities during activities of daily living. Gait Posture. 2010;31(3):300\u20136.","journal-title":"Gait Posture"},{"key":"1347_CR75","doi-asserted-by":"publisher","DOI":"10.3791\/57228","author":"M Alt Murphy","year":"2018","unstructured":"Alt Murphy M, Murphy S, Persson HC, Bergstr\u00f6m U-B, Sunnerhagen KS. Kinematic analysis using 3D motion capture of drinking task in people with and without upper-extremity impairments. J Vis Exp. 2018. https:\/\/doi.org\/10.3791\/57228.","journal-title":"J Vis Exp"},{"key":"1347_CR76","unstructured":"\u201cOpenCV: Camera Calibration.\u201d https:\/\/docs.opencv.org\/4.x\/dc\/dbb\/tutorial_py_calibration.html. Accessed 05 Aug 2024."},{"key":"1347_CR77","unstructured":"\u201cConfiguring stereo depth,\u201d Luxonis. https:\/\/docs.luxonis.com\/hardware\/platform\/depth\/configuring-stereo-depth. Accessed 18 Jul 2024."}],"container-title":["BioMedical Engineering OnLine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12938-025-01347-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12938-025-01347-y\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12938-025-01347-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,2,7]],"date-time":"2025-02-07T02:16:27Z","timestamp":1738894587000},"score":1,"resource":{"primary":{"URL":"https:\/\/biomedical-engineering-online.biomedcentral.com\/articles\/10.1186\/s12938-025-01347-y"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,2,7]]},"references-count":77,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["1347"],"URL":"https:\/\/doi.org\/10.1186\/s12938-025-01347-y","relation":{},"ISSN":["1475-925X"],"issn-type":[{"value":"1475-925X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,2,7]]},"assertion":[{"value":"27 September 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"29 January 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"7 February 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Original data collection for this study was approved by the Research Ethics Board of the University Health Network (23\u20135028), in compliance with the Canadian tri-council policy statement: ethical conduct for research involving humans. All participants signed written informed consent.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"The use of images in this article is consistent with the informed consent signed by the participants.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare no competing interests.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"12"}}