Autors: Tonchev K., Petkova, R. R., Manolova, A. H., Neshev S.
Title: Reconstruction of Animatable Human Body Model Using 3D Skeleton Data
Keywords: 3D Gaussian Splatting, Animatable Human Avatar

Abstract: The recent advances in 3D Gaussian Splatting opened the door for high quality 3D scene modeling and reconstruction with real-time performance. Reconstruction of human avatars and in particular animatable ones, is at the frontiers of the research in the recent couple of years, delivering better results in terms of performance and quality compared to the recent state of the art based on Neural Radiance Fields and implicit functions. In this work a novel approach to animatable 3D humans is proposed, using as driving input only the human skeleton, representing the target pose. The human model is based on 3D gaussians, and the corresponding transformation functions use Graph-Convolutional Neural Networks for predicting the required parameters. Evaluation results on popular dataset prove the proposed approach delivers state of the art results.

References

  1. W. Cheng, R. Chen, S. Fan, W. Yin, K. Chen, Z. Cai, and K. Y. Lin, "DNA-Rendering: A diverse neural actor repository for high-fidelity human-centric rendering", in Proc. IEEE/CVF Int. Conf. Comput. Vis. (ICCV), pp. 19982-19993, 2023.
  2. Z. Xiong, C. Li, K. Liu, H. Liao, J. Hu, J. Zhu, and X. Han, "MvHuman-Net: A large-scale dataset of multi-view daily dressing human captures", in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), pp. 19801-19811, 2024.
  3. A. Zubekhin, H. Zhu, P. Gotardo, T. Beeler, M. Habermann, and C. Theobalt, "GIGA: Generalizable sparse image-driven Gaussian avatars", arXiv preprint arXiv: 2504. 07144, 2025.
  4. B. Kerbl, G. Kopanas, T. Leimkühler, and G. Drettakis, "3D Gaussian splatting for real-time radiance field rendering", ACM Trans. Graph., vol. 42, no. 4, pp. 139-1, 2023.
  5. Z. Qian, S. Wang, M. Mihajlovic, A. Geiger, and S. Tang, "3DGSAvatar: Animatable avatars via deformable 3D Gaussian splatting", in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), pp. 5020-5030, 2024.
  6. S. Hu, T. Hu, and Z. Liu, "GauHuman: Articulated Gaussian splatting from monocular human videos", in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), pp. 20418-20431, 2024.
  7. A. Moreau, J. Song, H. Dhamo, R. Shaw, Y. Zhou, and E. Pérez-Pellitero, "Human Gaussian splatting: Real-time rendering of animatable avatars", in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), pp. 788-798, 2024.
  8. Y. Xu, K. Ye, T. Shao, and Y. Weng, "Animatable 3D Gaussians for modeling dynamic humans", Front. Comput. Sci., vol. 19, no. 9, pp. 1-11, 2025.
  9. M. Loper, N. Mahmood, J. Romero, G. Pons-Moll, and M. J. Black, "SMPL: A Skinned Multi-Person Linear Model", ACM Trans. Graph., vol. 34, no. 6, pp. 248: 1-248: 16, 2015.
  10. J. Wang, S. Tan, X. Zhen, S. Xu, F. Zheng, Z. He, and L. Shao, "Deep 3D human pose estimation: A review", Comput. Vis. Image Underst., vol. 210, p. 103225, 2021.
  11. V. Voleti, B. Oreshkin, F. Bocquelet, F. Harvey, L. S. Ménard, and C. Pal, "SMPL-IK: Learned morphology-aware inverse kinematics for AI driven artistic workflows", in SIGGRAPH Asia 2022 Technical Communications, pp. 1-7, 2022.
  12. Z. Zhang, "SMPL-IKS: An Inverse Kinematic Solver for 3D Human Mesh Recovery", Available at: Https: //dx. doi. org/10. 21227/w906-9h60, 2024.
  13. S. Wang, K. Schwarz, A. Geiger, and S. Tang, "Arah: Animatable volume rendering of articulated human SDFs", in Eur. Conf. Comput. Vis. (ECCV), pp. 1-19, Cham: Springer Nature Switzerland, 2022.
  14. B. Mildenhall, P. P. Srinivasan, M. Tancik, J. T. Barron, R. Ramamoorthi, and R. Ng, "NeRF: Representing scenes as neural radiance fields for view synthesis", Commun. ACM, vol. 65, no. 1, pp. 99-106, 2021.
  15. D. P. Kingma and J. Ba, "Adam: A method for stochastic optimization", aarXiv preprint arXiv: 1412. 6980, 2014.
  16. M. Li, S. Yao, C. Kai, Z. Xie, K. Chen, and Y. G. Jiang, "AniGaussian: Animatable Gaussian Avatar with Pose-guided Deformation", arXiv preprint arXiv: 2502. 19441, 2025.
  17. S. Peng, Y. Zhang, Y. Xu, Q. Wang, Q. Shuai, H. Bao, and X. Zhou, "Neural body: Implicit neural representations with structured latent codes for novel view synthesis of dynamic humans", in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), pp. 9054-9063, 2021.
  18. S. Peng, J. Dong, Q. Wang, S. Zhang, Q. Shuai, X. Zhou, and H. Bao, "Animatable neural radiance fields for modeling dynamic human bodies", in Proc. IEEE/CVF Int. Conf. Comput. Vis. (ICCV), pp. 14314-14323, 2021.

Issue

60th International Scientific Conference on Information, Communication and Energy Systems and Technologies, ICEST 2025 - Proceedings, 2025, Macedonia, https://doi.org/10.1109/ICEST66328.2025.11098291

Copyright IEEE

Вид: публикация в международен форум, публикация в реферирано издание, индексирана в Scopus