Vol. 3 No. 5
In Progress

Research Article

Digital Human Creation Based on Multi-View 3D Reconstruction: Usability Evaluation and Collaborative Workflow

Published 2026-09-30

  • Kun Fu
    • ,
  • Shuyao Shi
    • ,
  • Chengcheng Qi
    • ,
  • Xuwen Lan
    • ,
  • Pengfei Cai
    • ,
  • Zhen Liu

Kun Fu
Beijing Normal-Hong Kong Baptist University image/svg+xml

Shuyao Shi
Beijing Normal-Hong Kong Baptist University image/svg+xml

Chengcheng Qi
Beijing Normal-Hong Kong Baptist University image/svg+xml

Xuwen Lan
Beijing Normal-Hong Kong Baptist University image/svg+xml

Pengfei Cai
Yangcheng Evening News Group

Zhen Liu
Nanfang College Guangzhou image/svg+xml

Abstract

To create digital humans from actual people there is a practical workflow that includes capturing images, reconstructing models, fitting characters, animating faces and exchanging resources. This research has three parts; a preliminary usability testing of Metashape, creating characters from real people and developing a collaborative workflow. Six students in Digital Media Technology each had thirty minutes of instruction prior to being asked to complete seven reconstruction tasks on a 60 cm plaster bust of Voltaire. Of all possible (42) student/task combinations, thirty-four were successful which gave the study an overall task completion rate of eighty percent (80.95%). Four of those six students successfully completed the full reconstruction process resulting in a sixty-six percent (66.67%) full-process completion rate. On average, it took seventy-nine (78.75) minutes to complete the tasks. The majority of this time was spent on photographing the subject which accounted for fifty-nine percent (50.79%) of the time. Some common reasons for failures observed during the experiment included too little overlap between photographs, overly exposed areas and lost focus. An independent demonstration showed how a Metashape mesh model can be converted into a MetaHuman character that is rendered inside Unreal Engine 5. These results are used to develop a workflow that incorporates high, flat and low shooting rings for capturing images of subjects, HumanSeg as a method to remove backgrounds from images, cloud-based resource exchanges and Live Link Face as a means of providing real-time facial animation. The initial results suggest that providing guidance during image capture and checking image quality are key factors. The example of creating a character shows that creating such characters is feasible for certain use cases. Automation could improve several aspects including processing speed, visual fidelity and realism of facial movements but would need to be evaluated separately.

Keywords

Multi-view 3D reconstruction, Digital human, Usability, Photogrammetry, Metashape, MetaHuman, Collaborative workflow

PDF

References

  1. Amershi, S., Weld, D., Vorvoreanu, M., Fourney, A., Nushi, B., Collisson, P., Suh, J., Iqbal, S., Bennett, P. N., Inkpen, K., Teevan, J., Kikin-Gil, R., & Horvitz, E. (2019). Guidelines for human-AI interaction. Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems, 1–13. https://doi.org/10.1145/3290605.3300233 DOI: https://doi.org/10.1145/3290605.3300233
  2. Blanz, V., & Vetter, T. (1999). A morphable model for the synthesis of 3D faces. Proceedings of the 26th annual conference on Computer graphics and interactive techniques - SIGGRAPH '99, 187–194. https://doi.org/10.1145/311535.311556 DOI: https://doi.org/10.1145/311535.311556
  3. Cao, C., Weng, Y., Zhou, S., Tong, Y., & Zhou, K. (2014). FaceWarehouse: A 3D facial expression database for visual computing. IEEE Transactions on Visualization and Computer Graphics, 20(3), 413–425. https://doi.org/10.1109/tvcg.2013.249 DOI: https://doi.org/10.1109/TVCG.2013.249
  4. Cioffi, F., Masullo, M., Pascale, A., & Maffei, L. (2025). Speech intelligibility in virtual avatars: Comparison between audio and audio–visual-driven facial animation. Acoustics, 7(2), 30. https://doi.org/10.3390/acoustics7020030 DOI: https://doi.org/10.3390/acoustics7020030
  5. Fu, K., Li, P., Peng, X., Li, S., Zhao, Y., Chen, X., Ding, Y., & Ye, C. (2026). From technological closed-loop to human-machine trust: a systematic review of ethical and communication challenges of brain-computer interface in elderly care. Frontiers in Digital Health, 8, 1877688. https://doi.org/10.3389/fdgth.2026.1877688 DOI: https://doi.org/10.3389/fdgth.2026.1877688
  6. Fu, K., Ye, C., Wang, Z., Wu, M., Liu, Z., & Yuan, Y. (2025). Ethical dilemmas and the reconstruction of subjectivity in digital mourning in the age of AI: an empirical study on the acceptance intentions of bereaved family members of cancer patients. Frontiers in Digital Health, 7, 1618169. https://doi.org/10.3389/fdgth.2025.1618169 DOI: https://doi.org/10.3389/fdgth.2025.1618169
  7. Furukawa, Y., & Ponce, J. (2010). Accurate, dense, and robust multiview stereopsis. IEEE Transactions on Pattern Analysis and Machine Intelligence, 32(8), 1362–1376. https://doi.org/10.1109/tpami.2009.161 DOI: https://doi.org/10.1109/TPAMI.2009.161
  8. Li, T., Bolkart, T., Black, M. J., Li, H., & Romero, J. (2017). Learning a model of facial shape and expression from 4D scans. ACM Transactions on Graphics, 36(6), Article 194. https://doi.org/10.1145/3130800.3130813 DOI: https://doi.org/10.1145/3130800.3130813
  9. Lin, L., and Chen, J. (2022). Research on multi-view 3D face reconstruction methods. Video Engineering, (7), 203–207. https://doi.org/10.16280/j.videoe.2022.07.050. [In Chinese; title translated.]
  10. Lowe, D. G. (2004). Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision, 60(2), 91–110. https://doi.org/10.1023/b:visi.0000029664.99615.94 DOI: https://doi.org/10.1023/B:VISI.0000029664.99615.94
  11. Miao, D., and Wu, J. (2017). Image-based 3D reconstruction and accuracy assessment using Agisoft Photoscan. Engineering of Surveying and Mapping, (8), 41–44. https://doi.org/10.19349/j.cnki.issn1006-7949.2017.08.009. [In Chinese; title translated.]
  12. Mo, J., Chen, H., Ye, C., Wang, Z., & Chen, C. (2026). Exploring the drivers of users' adoption of museum digital humans. npj Heritage Science, 14, 43. https://doi.org/10.1038/s40494-026-02313-0 DOI: https://doi.org/10.1038/s40494-026-02313-0
  13. Mohren, J., & Schulze, M. (2024). Automated cleaning of tie point clouds following USGS guidelines in Agisoft Metashape professional (ver. 2.1.0). MethodsX, 12, 102679. https://doi.org/10.1016/j.mex.2024.102679 DOI: https://doi.org/10.1016/j.mex.2024.102679
  14. PaddlePaddle. (2024). PaddleSeg PP-HumanSeg documentation, release/2.6. https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.6/contrib/PP-HumanSeg. Accessed October 13, 2024.
  15. Remondino, F., Spera, M. G., Nocerino, E., Menna, F., & Nex, F. (2014). State of the art in high density image matching. The Photogrammetric Record, 29(146), 144–166. https://doi.org/10.1111/phor.12063 DOI: https://doi.org/10.1111/phor.12063
  16. Rivera, J., & Salinas, P. (2025). Workflow for photogrammetric digitization using Agisoft Metashape. Figshare. https://doi.org/10.1371/journal.pone.0336996.s003. DOI: https://doi.org/10.1371/journal.pone.0336996.s003
  17. Schönberger, J. L., & Frahm, J. M. (2016). Structure-from-motion revisited. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 4104–4113. https://doi.org/10.1109/cvpr.2016.445 DOI: https://doi.org/10.1109/CVPR.2016.445
  18. Schönberger, J. L., Zheng, E., Frahm, J. M., & Pollefeys, M. (2016). Pixelwise view selection for unstructured multi-view stereo. Computer Vision – ECCV 2016, Lecture Notes in Computer Science, 9907, 501–518. https://doi.org/10.1007/978-3-319-46487-9_31 DOI: https://doi.org/10.1007/978-3-319-46487-9_31
  19. Seitz, S., Curless, B., Diebel, J., Scharstein, D., & Szeliski, R. (2006). A comparison and evaluation of multi-view stereo reconstruction algorithms. 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1 (CVPR'06), 1, 519–528. https://doi.org/10.1109/cvpr.2006.19 DOI: https://doi.org/10.1109/CVPR.2006.19
  20. Sun, B. (2022). Multi-view 3D reconstruction based on deep learning. Master's thesis, Dalian University of Technology, Dalian, China. [In Chinese; title translated.]
  21. Sun, D. (2022). An introduction to MetaHuman principles and workflows. Zhihu. https://zhuanlan.zhihu.com/p/445646111. Accessed May 5, 2023. [In Chinese; title translated.]
  22. Wang, Y., and Liang, X. (2022). Virtual subjectivity and virtual materiality: On digital human representations. Journal of Fujian Normal University (Philosophy and Social Sciences Edition), (5), 49–58, 170. [In Chinese; title translated.]
  23. Xian, H. (2021). A study of Byung-Chul Han's aesthetic thought on digital humans. Master's thesis, Lanzhou University, Lanzhou, China. [In Chinese; title translated.]