Divide-and-Conquer: Dual-Hierarchical Optimization for Semantic 4D Gaussian Spatting
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917967856402432 |
|---|---|
| author | Yan, Zhiying Liang, Yiyuan Cai, Shilv Zhang, Tao Zhong, Sheng Yan, Luxin Zou, Xu |
| author_facet | Yan, Zhiying Liang, Yiyuan Cai, Shilv Zhang, Tao Zhong, Sheng Yan, Luxin Zou, Xu |
| contents | Semantic 4D Gaussians can be used for reconstructing and understanding dynamic scenes, with temporal variations than static scenes. Directly applying static methods to understand dynamic scenes will fail to capture the temporal features. Few works focus on dynamic scene understanding based on Gaussian Splatting, since once the same update strategy is employed for both dynamic and static parts, regardless of the distinction and interaction between Gaussians, significant artifacts and noise appear. We propose Dual-Hierarchical Optimization (DHO), which consists of Hierarchical Gaussian Flow and Hierarchical Gaussian Guidance in a divide-and-conquer manner. The former implements effective division of static and dynamic rendering and features. The latter helps to mitigate the issue of dynamic foreground rendering distortion in textured complex scenes. Extensive experiments show that our method consistently outperforms the baselines on both synthetic and real-world datasets, and supports various downstream tasks. Project Page: https://sweety-yan.github.io/DHO. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_19332 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Divide-and-Conquer: Dual-Hierarchical Optimization for Semantic 4D Gaussian Spatting Yan, Zhiying Liang, Yiyuan Cai, Shilv Zhang, Tao Zhong, Sheng Yan, Luxin Zou, Xu Computer Vision and Pattern Recognition Semantic 4D Gaussians can be used for reconstructing and understanding dynamic scenes, with temporal variations than static scenes. Directly applying static methods to understand dynamic scenes will fail to capture the temporal features. Few works focus on dynamic scene understanding based on Gaussian Splatting, since once the same update strategy is employed for both dynamic and static parts, regardless of the distinction and interaction between Gaussians, significant artifacts and noise appear. We propose Dual-Hierarchical Optimization (DHO), which consists of Hierarchical Gaussian Flow and Hierarchical Gaussian Guidance in a divide-and-conquer manner. The former implements effective division of static and dynamic rendering and features. The latter helps to mitigate the issue of dynamic foreground rendering distortion in textured complex scenes. Extensive experiments show that our method consistently outperforms the baselines on both synthetic and real-world datasets, and supports various downstream tasks. Project Page: https://sweety-yan.github.io/DHO. |
| title | Divide-and-Conquer: Dual-Hierarchical Optimization for Semantic 4D Gaussian Spatting |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2503.19332 |