{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,31]],"date-time":"2026-10-31T02:21:57Z","timestamp":1793413317767,"version":"4.5.0"},"reference-count":40,"publisher":"Public Library of Science (PLoS)","issue":"9","license":[{"start":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T00:00:00Z","timestamp":1790553600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Young Scientific and Technological Talents Program for Higher Education Institutions of Inner Mongolia","award":["NJYT24061"],"award-info":[{"award-number":["NJYT24061"]}]},{"DOI":"10.13039\/100020593","name":"Fundamental Research Funds for the Central Universities of Beijing University of Chemical Technology","doi-asserted-by":"publisher","award":["JY20220249"],"award-info":[{"award-number":["JY20220249"]}],"id":[{"id":"10.13039\/100020593","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Monocular 3D Human Pose Estimation (3D HPE) is a core task in computer vision, which provides critical technical support for applications including VR\/AR, immersive human-computer interaction, digital human animation, as well as sports motion capture, athletic technique diagnosis, physical fitness assessment and sports rehabilitation training. However, existing methods still face three key limitations: diffusion-based methods lack explicit human anatomical prior constraints in the diffusion process; disentangled-based methods suffer from severe hierarchical error accumulation along the skeletal kinematic tree; and most mainstream methods fail to fully mine the fine-grained hierarchical spatio-temporal correlations of the human skeleton, leading to increased estimation error for high-hierarchy joints. To address these challenges, we propose a novel Hierarchy-Aware Disentangled Diffusion framework with Spatio-Temporal Denoising (HADD) for monocular 3D HPE, which deeply integrates disentanglement strategy with the diffusion model and embeds skeletal hierarchical information into the full diffusion pipeline. In the forward diffusion process, HADD disentangles 3D pose into bone length and bone direction features, and injects Gaussian noise into the two features separately. In the reverse denoising process, we design a Hierarchical Spatial and Temporal Denoising (HSTD) module to accurately model the hierarchical spatio-temporal relationships of the human skeleton and alleviate error amplification of high-hierarchy joints. Meanwhile, a hybrid loss function combining 3D disentanglement loss and 3D pose loss is constructed to realize dual supervision at the bone and joint levels. Extensive experiments on Human3.6M and MPI-INF-3DHP datasets show that HADD achieves consistent performance improvement over state-of-the-art methods. On Human3.6M, it reduces average MPJPE by up to 12.9% compared with traditional disentangled-based methods, and outperforms mainstream diffusion-based probabilistic methods under lightweight inference configurations. On MPI-INF-3DHP, HADD achieves an MPJPE of 29.2 mm, a PCK of 98.5%, and an AUC of 78.1 under single-hypothesis inference, reaching new state-of-the-art performance on this dataset under the single-hypothesis inference setting with only ground-truth 2D joint sequences as input. This work provides an effective solution to the core limitations of existing 3D HPE methods, and also offers reliable technical support for fine-grained motion analysis and quantitative assessment in sports training and rehabilitation scenarios.<\/jats:p>","DOI":"10.1371\/journal.pone.0359321","type":"journal-article","created":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T17:50:32Z","timestamp":1790617832000},"page":"e0359321","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":1,"title":["HADD: A hierarchy-aware disentangled diffusion framework with spatio-temporal denoising for 3D human pose estimation"],"prefix":"10.1371","volume":"21","author":[{"given":"Jinze","family":"Jiang","sequence":"first","affiliation":[{"name":"Fujian Agriculture and Forestry University","place":["Fuzhou, China"],"department":["Department of Physical Education"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-1300-6294","authenticated-orcid":true,"given":"Zhongheng","family":"Jian","sequence":"additional","affiliation":[{"name":"Inner Mongolia University of Technology","place":["Huhhot, China"],"department":["College of Intelligent Science and Technology"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hualing","family":"Zheng","sequence":"additional","affiliation":[{"name":"Fujian University of Technology","place":["Fuzhou, China"],"department":["Department of Physical Education"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Minsi","family":"Liang","sequence":"additional","affiliation":[{"name":"Nanyang Technological University","place":["Singapore, Singapore"],"department":["School of Electrical and Electronic Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuxing","family":"Shi","sequence":"additional","affiliation":[{"name":"Sichuan Polytechnic University","place":["Deyang, China"],"department":["School of Software Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liming","family":"Zhang","sequence":"additional","affiliation":[{"name":"Hebei Petroleum University of Technology","place":["Chengde, China"],"department":["Department of Mechanical Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,9,28]]},"reference":[{"key":"pone.0359321.ref001","first-page":"947","article-title":"Mixed Reality 6D Object Pose Estimation using Deep Learning for Visual Markerless Surgical Navigation.","volume-title":"IEEE Conference on Virtual Reality and 3D User Interfaces, VR 2025 - Abstracts and Workshops, Saint Malo, France, March 8-12, 2025","author":"M Trojak","year":"2025"},{"issue":"20","key":"pone.0359321.ref002","doi-asserted-by":"crossref","first-page":"10947","DOI":"10.1007\/s00500-022-07240-3","article-title":"Key technologies of human\u2013computer interaction for immersive somatosensory interactive games using VR technology","volume":"26","author":"P Gao","year":"2022","journal-title":"Soft Comput"},{"issue":"7","key":"pone.0359321.ref003","doi-asserted-by":"crossref","first-page":"60","DOI":"10.3390\/mti8070060","article-title":"A virtual reality direct-manipulation tool for posing and animation of digital human bodies: An evaluation of creativity support","volume":"8","author":"Y Benbelkheir","year":"2024","journal-title":"Multimodal Technol Interact"},{"key":"pone.0359321.ref004","first-page":"1","article-title":"Performance Evaluations of IEEE 802.11ad and Human Pose Detection towards Intelligent Video Surveillance System.","volume-title":"IEEE International Conference on Consumer Electronics - Taiwan, ICCE-TW 2019, Yilan, Taiwan, May 20-22, 2019","author":"M Okano","year":"2019"},{"key":"pone.0359321.ref005","first-page":"1","article-title":"Using OpenPose for Enhancing CPR Training: A Data-Driven Approach.","volume-title":"IEEE Symposium on Computers and Communications, ISCC 2025, Bologna, Italy, July 2-5, 2025","author":"S Mwali","year":"2025"},{"key":"pone.0359321.ref006","doi-asserted-by":"crossref","DOI":"10.7717\/peerj-cs.2933","article-title":"Automatic detection of teacher behavior in classroom videos using AlphaPose and Faster R-CNN algorithms","volume":"11","author":"J Huang","year":"2025","journal-title":"PeerJ Comput Sci"},{"key":"pone.0359321.ref007","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TIM.2025.3551025","article-title":"Spatial\u2013temporal\u2013geometric graph convolutional network for 3-D human pose estimation from multiview video","volume":"74","author":"K Dong","year":"2025","journal-title":"IEEE Trans Instrum Meas"},{"key":"pone.0359321.ref008","doi-asserted-by":"crossref","first-page":"104696","DOI":"10.1016\/j.jvcir.2025.104696","article-title":"3D human pose estimation based on a Hybrid approach of Transformer and GCN-Former","volume":"115","author":"X Pan","year":"2026","journal-title":"J Visual Commun Image Represent"},{"issue":"2","key":"pone.0359321.ref009","doi-asserted-by":"crossref","first-page":"143","DOI":"10.1007\/s00371-025-04313-w","article-title":"Enhancing human pose estimation accuracy with pyramid fusion vision transformers","volume":"42","author":"S Bian","year":"2026","journal-title":"Visual Computer"},{"issue":"1","key":"pone.0359321.ref010","doi-asserted-by":"crossref","first-page":"32","DOI":"10.1007\/s10462-025-11430-4","article-title":"A survey on deep learning for 2D and 3D human pose estimation","volume":"59","author":"MM Kappan","year":"2026","journal-title":"Artif Intell Rev"},{"key":"pone.0359321.ref011","doi-asserted-by":"crossref","first-page":"79118","DOI":"10.1109\/ACCESS.2025.3566109","article-title":"Toward efficient generalization in 3D human pose estimation via a canonical domain approach","volume":"13","author":"H Lee","year":"2025","journal-title":"IEEE Access"},{"issue":"02","key":"pone.0359321.ref012","article-title":"3D hand pose estimation based on hand denoising diffusion probabilistic robustness model","volume":"40","author":"L She","year":"2025","journal-title":"Int J Patt Recogn Artif Intell"},{"key":"pone.0359321.ref013","doi-asserted-by":"crossref","first-page":"73319","DOI":"10.1109\/ACCESS.2025.3563898","article-title":"DCT-DiffPose: A lightweight diffusion model with multi-hypothesis for 3D human pose estimation","volume":"13","author":"L Zhong","year":"2025","journal-title":"IEEE Access"},{"issue":"7","key":"pone.0359321.ref014","article-title":"Text\u2010guided diffusion with spectral convolution for 3D human pose estimation","volume":"44","author":"L Shi","year":"2025","journal-title":"Comput Graph Forum"},{"key":"pone.0359321.ref015","doi-asserted-by":"crossref","first-page":"92712","DOI":"10.1109\/ACCESS.2025.3571760","article-title":"AMT-Net: Adversarial motion transfer network with disentangled shape and pose for realistic image animation","volume":"13","author":"N Asebe Teka","year":"2025","journal-title":"IEEE Access"},{"issue":"8","key":"pone.0359321.ref016","doi-asserted-by":"crossref","DOI":"10.1007\/s44443-025-00262-5","article-title":"D-NeuRA: customizable dynamic neural rendering for human avatars with disentangled pose and appearance","volume":"37","author":"M Jiang","year":"2025","journal-title":"J King Saud Univ Comput Inf Sci"},{"key":"pone.0359321.ref017","first-page":"1189","article-title":"Demo: FreePose: Real-Time View-Invariant 3D Human Pose Estimation via Motion-View Disentanglement.","volume-title":"Proceedings of the 31st Annual International Conference on Mobile Computing and Networking, ACM MOBICOM 2025, Hong Kong, November 4-8, 2025","author":"Y Liu","year":"2025"},{"issue":"1","key":"pone.0359321.ref018","doi-asserted-by":"crossref","first-page":"198","DOI":"10.1109\/TCSVT.2021.3057267","article-title":"Anatomy-aware 3D human pose estimation with bone-based pose decomposition","volume":"32","author":"T Chen","year":"2022","journal-title":"IEEE Trans Circuits Syst Video Technol"},{"key":"pone.0359321.ref019","first-page":"13222","article-title":"MixSTE: Seq2seq Mixed Spatio-Temporal Encoder for 3D Human Pose Estimation in Video.","volume-title":"IEEE\/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022, New Orleans, LA, USA, June 18-24, 2022","author":"J Zhang","year":"2022"},{"key":"pone.0359321.ref020","first-page":"14715","article-title":"Diffusion-Based 3D Human Pose Estimation with Multi-Hypothesis Aggregation.","volume-title":"IEEE\/CVF International Conference on Computer Vision, ICCV 2023, Paris, France, October 1-6, 2023","author":"W Shan","year":"2023"},{"key":"pone.0359321.ref021","doi-asserted-by":"crossref","first-page":"159050","DOI":"10.1109\/ACCESS.2025.3607702","article-title":"PRAMGCN-Net: 3D human pose estimation with a parameterized routing adjacency modulation graph convolutional network","volume":"13","author":"A Pramono","year":"2025","journal-title":"IEEE Access"},{"key":"pone.0359321.ref022","doi-asserted-by":"crossref","DOI":"10.7717\/peerj-cs.2907","article-title":"Yoga pose recognition using dual structure convolutional neural network","volume":"11","author":"X Meng","year":"2025","journal-title":"PeerJ Comput Sci"},{"issue":"1","key":"pone.0359321.ref023","doi-asserted-by":"crossref","first-page":"869","DOI":"10.1109\/JIOT.2024.3470751","article-title":"Hierarchical local temporal network for 2D-to-3D human pose estimation","volume":"12","author":"X Yan","year":"2025","journal-title":"IEEE Internet Things J"},{"issue":"10","key":"pone.0359321.ref024","article-title":"Whole-body 2D human pose estimation based on human keypoints distribution constraint and adaptive Gaussian factor","volume":"11","author":"F Wei","year":"2025","journal-title":"Complex Intell Syst"},{"key":"pone.0359321.ref025","first-page":"11636","article-title":"3D Human Pose Estimation with Spatial and Temporal Transformers.","volume-title":"2021 IEEE\/CVF International Conference on Computer Vision, ICCV 2021, Montreal, QC, Canada, October 10-17, 2021","author":"C Zheng","year":"2021"},{"key":"pone.0359321.ref026","first-page":"8877","article-title":"PoseFormerV2: Exploring Frequency Domain for Efficient and Robust 3D Human Pose Estimation.","volume-title":"IEEE\/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023, Vancouver, BC, Canada, June 17-24, 2023","author":"Q Zhao","year":"2023"},{"key":"pone.0359321.ref027","first-page":"896","article-title":"Deep Kinematics Analysis for Monocular 3D Human Pose Estimation.","volume-title":"2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2020, Seattle, WA, USA, June 13-19, 2020. Computer Vision Foundation","author":"J Xu","year":"2020"},{"key":"pone.0359321.ref028","unstructured":"Wang G, Zeng H, Wang Z, Liu Z, Wang H. Motion Projection Consistency Based 3D Human Pose Estimation with Virtual Bones from Monocular Videos. CoRR. 2021. http:\/\/arxiv.org\/abs\/2106.14706"},{"key":"pone.0359321.ref029","unstructured":"Ho J, Jain A, Abbeel P. Denoising Diffusion Probabilistic Models. In: Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual; 2020."},{"key":"pone.0359321.ref030","article-title":"Denoising Diffusion Implicit Models.","volume-title":"9th International Conference on Learning Representations, ICLR 2021, Virtual Event, Austria, May 3-7, 2021","author":"J Song","year":"2021"},{"key":"pone.0359321.ref031","doi-asserted-by":"crossref","first-page":"130306","DOI":"10.1016\/j.neucom.2025.130306","article-title":"Uncertainty-guided diffusion model for 3D human pose estimation","volume":"641","author":"Z Liu","year":"2025","journal-title":"Neurocomputing"},{"issue":"1","key":"pone.0359321.ref032","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1007\/s00530-024-01569-5","article-title":"3d human pose estimation based on conditional dual-branch diffusion","volume":"31","author":"J Li","year":"2025","journal-title":"Multim Syst"},{"key":"pone.0359321.ref033","first-page":"4800","article-title":"GFPose: Learning 3D Human Pose Prior with Gradient Fields.","volume-title":"IEEE\/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023, Vancouver, BC, Canada, June 17-24, 2023","author":"H Ci","year":"2023"},{"key":"pone.0359321.ref034","first-page":"13041","article-title":"DiffPose: Toward More Reliable 3D Pose Estimation.","volume-title":"IEEE\/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023, Vancouver, BC, Canada, June 17-24, 2023","author":"J Gong","year":"2023"},{"issue":"7","key":"pone.0359321.ref035","doi-asserted-by":"crossref","first-page":"1325","DOI":"10.1109\/TPAMI.2013.248","article-title":"Human3.6M: Large Scale Datasets and Predictive Methods for 3D Human Sensing in Natural Environments","volume":"36","author":"C Ionescu","year":"2014","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"pone.0359321.ref036","first-page":"461","article-title":"P-STMO: Pre-trained Spatial Temporal Many-to-One Model for 3D Human Pose Estimation.","volume-title":"Computer Vision - ECCV 2022 - 17th European Conference, Tel Aviv, Israel, October 23-27, 2022, Proceedings, Part V. vol. 13665 of Lecture Notes in Computer Science","author":"W Shan","year":"2022"},{"issue":"2","key":"pone.0359321.ref037","doi-asserted-by":"crossref","first-page":"784","DOI":"10.1109\/TCDS.2022.3185146","article-title":"Motion projection consistency-based 3-D human pose estimation with virtual bones from monocular videos","volume":"15","author":"G Wang","year":"2023","journal-title":"IEEE Trans Cogn Dev Syst"},{"key":"pone.0359321.ref038","first-page":"4790","article-title":"3D Human Pose Estimation with Spatio-Temporal Criss-Cross Attention.","volume-title":"IEEE\/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023, Vancouver, BC, Canada, June 17-24, 2023","author":"Z Tang","year":"2023"},{"key":"pone.0359321.ref039","doi-asserted-by":"crossref","first-page":"112507","DOI":"10.1016\/j.patcog.2025.112507","article-title":"Spectral compression transformer with line pose graph for monocular 3D human pose estimation","volume":"172","author":"Z Zheng","year":"2026","journal-title":"Pattern Recognit"},{"key":"pone.0359321.ref040","first-page":"13137","article-title":"MHFormer: Multi-Hypothesis Transformer for 3D Human Pose Estimation.","volume-title":"IEEE\/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022, New Orleans, LA, USA, June 18-24, 2022","author":"W Li","year":"2022"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0359321","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T17:50:46Z","timestamp":1790617846000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0359321"}},"subtitle":[],"editor":[{"given":"Gennady S.","family":"Cymbalyuk","sequence":"first","affiliation":[{"name":"Georgia State University","place":["UNITED STATES OF AMERICA"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,9,28]]},"references-count":40,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,9,28]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0359321","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,28]]}}}