4D Gaussian Splatting for Dynamic Scene Representation
Main Article Content
Abstract
In this paper, we present a novel approach for dynamic scene representation using 4D Gaussian splatting, extending traditional 3D Gaussian splatting into the temporal domain. Our method models scenes as a collection of Gaussian kernels parameterized over space and time, enabling continuous and smooth interpolation of dynamic content. This formulation allows efficient rendering and compact representation of temporal variations without relying on discrete frame storage. We validate our approach on a synthetic toy dataset featuring a moving Gaussian object in 3D space over time. Experimental results demonstrate accurate reconstruction, real-time rendering capabilities, and temporally coherent outputs compared to baseline frame-based methods. The proposed 4D Gaussian splatting offers a promising direction for efficient and interpretable dynamic scene modeling.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0). To view a copy of this license, visit https://creativecommons.org/licenses/by/4.0/
References
Qi-Yuan Feng, Hao-Xiang Chen, Qun-Ce Xu, and Tai-Jiang Mu. Sls4d: sparse latent space for 4d novel view synthesis. IEEE Transactions on Visualization and Computer Graphics, 2024.
Wanshui Gan, Hongbin Xu, Yi Huang, Shifeng Chen, and Naoto Yokoya. V4d: Voxel for 4d novel view synthesis. IEEE Transactions on Visualization and Computer Graphics, 30(2):1579–1591, 2023.
Moritz Kappel, Florian Hahlbohm, Timon Scholz, Susana Castillo, Christian Theobalt, Martin Eisemann, Vladislav Golyanik, and Marcus Magnor. D-npc: Dynamic neural point clouds for non-rigid view synthesis from monocular video. In Computer Graphics Forum, page e70038. Wiley Online Library, 2025.
Bernhard Kerbl, Georgios Kopanas, Thomas Leimku ̈hler, and George Drettakis. 3d gaussian splatting for real-time radiance field rendering. ACM Trans. Graph., 42(4):139–1, 2023.
Lingzhi Li, Zhen Shen, Zhongshu Wang, Li Shen, and Ping Tan. Streaming radiance fields for 3d video synthesis. Advances in Neural Information Processing Systems, 35:13485–13498, 2022.
Zhengqi Li, Simon Niklaus, Noah Snavely, and Oliver Wang. Neural scene flow fields for space-time view synthesis of dynamic scenes. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 6498–6508, 2021.
Yu-Lun Liu, Chen Gao, Andreas Meuleman, Hung-Yu Tseng, Ayush Saraf, Changil Kim, Yung-Yu Chuang, Johannes Kopf, and Jia-Bin Huang. Robust dynamic radiance fields. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 13–23, 2023.
Keunhong Park, Utkarsh Sinha, Jonathan T Barron, Sofien Bouaziz, Dan B Goldman, Steven M Seitz, and Ricardo Martin-Brualla. Nerfies: Deformable neural radiance fields. In Proceedings of the IEEE/CVF international conference on computer vision, pages 5865– 5874, 2021.
Albert Pumarola, Enric Corona, Gerard Pons-Moll, and Francesc Moreno-Noguer. D-nerf: Neural radiance fields for dynamic scenes. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10318–10327, 2021.
Ruizhi Shao, Zerong Zheng, Hanzhang Tu, Boning Liu, Hongwen Zhang, and Yebin Liu. Tensor4d: Efficient neural 4d decomposition for high-fidelity dynamic reconstruction and rendering. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 16632–16642, 2023.
Wenqi Xian, Jia-Bin Huang, Johannes Kopf, and Changil Kim. Space-time neural irradiance fields for free-viewpoint video. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 9421–9431, 2021.