Abstract
To address the limitations of existing 3D human body reconstruction methods in terms of insufficient precision and coarse detail construction, this paper proposes an optimized solution integrating temporal information, human behavior recognition, and multi-module collaboration. The algorithm centers on pixel-aligned hidden functions to establish a full-process reconstruction framework encompassing "temporal constraints-behavioral guidance-feature selection-attitude optimization-3D mapping-detail enhancement". By employing the SAD algorithm for optimal matching point selection and LSTM for temporal dependency capture, the algorithm achieves cross-frame feature coordination. Subsequently, the CNN-LSTM model performs human behavior recognition, using behavioral categories to guide SMPL model’s attitude parameter prior and attitude discriminator constraints. Posture normalization eliminates individual variations, while the integration of SMPL model and PIFuHD hidden function enables structured 3D mapping. Finally, octree acceleration grids are utilized to output high-precision 3D human models. Experimental results demonstrate that the proposed algorithm outperforms traditional and literature methods, achieving stable human detail construction in both static standing scenarios and dynamic running scenes.
Data availability
The datasets used and/or analyzed during the current study are available from the corresponding author Feng Wang on reasonable request via e-mail iswf@xhsysu.edu.cn.
References
Chen, Ange, Chengdong, Wu. & Leng, Chuanjiang. Hourglass-GCN for 3D human pose estimation using skeleton structure and view correlation[J]. Comput. Mater. Continua 82(1), 173–191 (2025).
Ronghui, Wu. Stretchable spring-sheathed yarn sensor for 3D dynamic body reconstruction assisted by transfer learning[J]. InfoMat 6(4), 109–123 (2024).
Zhang, Fenghao et al. 3D hand pose and shape estimation from monocular RGB via efficient 2D cues[J]. Comput. Visual Media 10(1), 79–96 (2024).
Cai, X. et al. Single depth image 3D face reconstruction via domain adaptive learning. Front. Comput. Sci. 18(1), 181342 (2024).
Lin, J. et al. Overview of 3D human pose estimation[J]. Comput. Model. Eng. Sci. 3, 1621–1651 (2023).
Hur, J. S., Lee, H. G., Kang, S., Yeo, Y. C. & Kim, S. K. 3D Face reconstruction from a single image using a combined PCA-LPP method. Comput. Mater. Continua 74(3), 6213 (2023).
Kang, J., Fan, W., Li, Y., Liu, R. & Zhou, D. 3D human pose estimation using two-stream architecture with joint training[J]. Comput. Modeling Eng. Sci. 10, 607–629 (2023).
Zhou, Mi., Liu, Rui, Yi, Pengfei & Zhou, Dongsheng. ER-Net: Efficient recalibration network for multi-view multi-person 3D pose estimation[J]. Comput. Modeling Eng. Sci. 8, 2093–2109 (2023).
Yu, C. et al. RFPose-OT: RF-based 3D human pose estimation via optimal transport theory. Front. Inform. Technol. Electron. Eng. 24(10), 1445–1457 (2023).
Li, C., Liu, S., Yao, L. & Zou, S. Video-based body geometric aware network for 3D human pose estimation. Optoelectron. Lett. 18(5), 313–320 (2022).
Yang, Chao, Wang, Xuyu & Mao, Shiwen. RFID-based 3D human pose tracking: a subject generalization approach. Digital Commun. Netw. 8(3), 278–288 (2022).
Zeng, Xiaoxing, Wu, Zhelun, Peng, Xiaojiang & Qiao, Yu. Joint 3D facial shape reconstruction and texture completion from a single image. Computational Visual Media 8(2), 239–256 (2022).
Tang, Yanlong et al. 3D corrective nose reconstruction from a single image. Computational Visual Media 8(2), 225–237 (2022).
Jin, Peng et al. Weakly-supervised single-view dense 3D point cloud reconstruction via differentiable renderer. Chin. J. Mech. Eng. 34(5), 195–205 (2021).
Khan, Khalil et al. 3D head pose estimation through facial features and deep convolutional neural networks. Comput. Mater. Continua 2, 1757–1770 (2021).
Li, Xianhua, Yu, Haohao, Tian, Shuoyu, Lin, Fengtao & Masood, Usama. Multi-branch high-dimensional guided transformer-based 3D human posture estimation. Comput. Mater. Continua 78(3), 3551–3564 (2024).
He, Qiqi et al. From digital human modeling to human digital twin: framework and perspectives in human factors. Chin. J. Mech. Eng. 37(1), 1–14 (2024).
Ellouze, Ameni, Kadri, Nesrine, Alaerjan, Alaa & Ksantini, Mohamed. Combined CNN-LSTM deep learning algorithms for recognizing human physical activities in large and distributed manners:a recommendation system. Comput. Mater. Continua 79(4), 351–372 (2024).
Muthunambu, Navaneetha Krishnan et al. A novel eccentric intrusion detection model based on recurrent neural networks with leveraging LSTM. Comput. Mater. Continua 78(3), 3089–3127 (2024).
Lai, Yi-Chun., Chiang, Shu-Yin., Kan, Yao-Chiang. & Lin, Hsueh-Chun. Coupling analysis of multiple machine learning models for human activity recognition. Comput. Mater. Continua 79(6), 3783–3803 (2024).
Shen, Jiajie, Wang, Yan & Zhang, Dongxu. A novel locomotion rule embedding long short-term memory network with attention for human locomotor intent classification using multi-sensors signals. Comput. Mater. Continua 79(6), 4349–4370 (2024).
Zhu, Meng et al. sEMG-based lower limb motion prediction using CNN-LSTM with improved PCA optimization algorithm. J. Bionic Eng. 20(2), 612–627 (2023).
Almagrabi, Alaa Omran. A deep CNN-LSTM-based feature extraction for cyber-physical system monitoring. Comput. Mater. Continua 76(8), 2079–2093 (2023).
Liu, Pengpeng, Zhang, Guixuan, Zhang, Shuwu, Li, Yuanhao & Zeng, Zhi. Skeleton-aware implicit function for single-view human reconstruction. CAAI Trans. Intell. Technol. 8(2), 379–389 (2023).
Chen, Lu., Peng, Sida & Zhou, Xiaowei. Towards efficient and photorealistic 3D human reconstruction: a brief survey. Visual Informatics 5(4), 11–19 (2021).
Zhang, Jing, Yu, Keping, Wen, Zheng, Qi, Xin & Paul, Anup Kumar. 3D reconstruction for motion blurred images using deep learning-based intelligent systems. Comput. Mater. Continua 2, 2087–2104 (2021).
Zhou, Xuhong, Li, Shuai, Liu, Jiepeng, Wu, Zhou & Chen, Yohchia Frank. Construction activity analysis of workers based on human posture estimation information. Engineering 33(2), 225–236 (2024).
Hanif, Ch Avais et al. Human Gait recognition for biometrics application based on deep learning fusion assisted framework. Comput. Mater. Continua 78(1), 357–374 (2024).
Zhang, Tao & Cao, Yi. Improved lightweight deep learning algorithm in 3D reconstruction. Comput. Mater. Contin. 9, 5315–5325 (2022).
Author information
Authors and Affiliations
Contributions
Zemin Qiu: Conceptualization, methodology, software, validation, formal analysis, investigation, resources, data curation, writing—original draft preparation Jiajun Zou: methodology, software, validation, formal analysis Shaojiang Liu: formal analysis, investigation, resources Feng Wang: writing—review and editing, visualization, supervision, project administration, funding acquisition
Fundings
This research was supported by the 2024 University-level Research Project of Guangzhou Xinhua University (No.2024KYZDZK02), the Guangdong Province Key Construction Discipline Research Capacity Enhancement Project (Nos. 2021ZDJS144,2024ZDJS130), the Characteristic Innovation Category Project of Guangdong Ordinary Colleges and Universities (No. 2024KTSCX127), the School-level Scientific Research Project of Guangzhou Xinhua University (No. 2024KYCXTD02), the Young Innovative Talents Category Project of Guangdong Ordinary Colleges and Universities (Nos. 2023KQNCX124, 2024KQNCX076), and the Key Research Platforms and Projects of Regular Higher Education Institutions in Guangdong Province (No.2025ZDZX3059).
Corresponding author
Ethics declarations
Competing interest
The authors declare no competing financial or non-financial interests.
Ethics
All datasets and images involved in this study are non-confidential and can be made public. Upon approval by the Institutional Review Board and with reasonable justification, reasonable data access requests can be made to the corresponding author.
Informed consent
All participants were fully informed of the study purpose, procedures, potential risks, and benefits, and provided written informed consent in accordance with the Declaration of Helsinki. No coercion or inducement was used to obtain consent, and participants retained the right to withdraw from the study at any time without prejudice.
Compliance with guidelines and regulations
The study was conducted in accordance with the Declaration of Helsinki, the studies involving human participants were reviewed and approved by School of Information and Intelligence Engineering, Guangzhou Xinhua University Ethics Committee (Approval Number: 21023022023). The participants provided their written informed consent to participate in this study. All methods were performed in accordance with relevant guidelines and regulations.
Author agreement for publication
All authors have read and approved the final version of the manuscript, and consent to its publication in the journal.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Qiu, Z., Zou, J., Liu, S. et al. Multi-module collaborative 3D human body modeling algorithm based on PIFuHD. Sci Rep (2026). https://doi.org/10.1038/s41598-026-46008-9
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-46008-9
