Salvato in:
| Autori principali: | , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2409.06685 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909310870618112 |
|---|---|
| author | Chen, Junyi Ye, Weicai Wang, Yifan Chen, Danpeng Huang, Di Ouyang, Wanli Zhang, Guofeng Qiao, Yu He, Tong |
| author_facet | Chen, Junyi Ye, Weicai Wang, Yifan Chen, Danpeng Huang, Di Ouyang, Wanli Zhang, Guofeng Qiao, Yu He, Tong |
| contents | 3D Gaussian Splatting (3DGS) has shown promising performance in novel view synthesis. Previous methods adapt it to obtaining surfaces of either individual 3D objects or within limited scenes. In this paper, we make the first attempt to tackle the challenging task of large-scale scene surface reconstruction. This task is particularly difficult due to the high GPU memory consumption, different levels of details for geometric representation, and noticeable inconsistencies in appearance. To this end, we propose GigaGS, the first work for high-quality surface reconstruction for large-scale scenes using 3DGS. GigaGS first applies a partitioning strategy based on the mutual visibility of spatial regions, which effectively grouping cameras for parallel processing. To enhance the quality of the surface, we also propose novel multi-view photometric and geometric consistency constraints based on Level-of-Detail representation. In doing so, our method can reconstruct detailed surface structures. Comprehensive experiments are conducted on various datasets. The consistent improvement demonstrates the superiority of GigaGS. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_06685 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | GigaGS: Scaling up Planar-Based 3D Gaussians for Large Scene Surface Reconstruction Chen, Junyi Ye, Weicai Wang, Yifan Chen, Danpeng Huang, Di Ouyang, Wanli Zhang, Guofeng Qiao, Yu He, Tong Computer Vision and Pattern Recognition 3D Gaussian Splatting (3DGS) has shown promising performance in novel view synthesis. Previous methods adapt it to obtaining surfaces of either individual 3D objects or within limited scenes. In this paper, we make the first attempt to tackle the challenging task of large-scale scene surface reconstruction. This task is particularly difficult due to the high GPU memory consumption, different levels of details for geometric representation, and noticeable inconsistencies in appearance. To this end, we propose GigaGS, the first work for high-quality surface reconstruction for large-scale scenes using 3DGS. GigaGS first applies a partitioning strategy based on the mutual visibility of spatial regions, which effectively grouping cameras for parallel processing. To enhance the quality of the surface, we also propose novel multi-view photometric and geometric consistency constraints based on Level-of-Detail representation. In doing so, our method can reconstruct detailed surface structures. Comprehensive experiments are conducted on various datasets. The consistent improvement demonstrates the superiority of GigaGS. |
| title | GigaGS: Scaling up Planar-Based 3D Gaussians for Large Scene Surface Reconstruction |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2409.06685 |