Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.15677 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918144255197184 |
|---|---|
| author | Lee, Gahye Kim, Hyomin Ju, Gwangjin Son, Jooeun Yoon, Hyejeong Lee, Seungyong |
| author_facet | Lee, Gahye Kim, Hyomin Ju, Gwangjin Son, Jooeun Yoon, Hyejeong Lee, Seungyong |
| contents | We propose Camera Splatting, a novel view optimization framework for novel view synthesis. Each camera is modeled as a 3D Gaussian, referred to as a camera splat, and virtual cameras, termed point cameras, are placed at 3D points sampled near the surface to observe the distribution of camera splats. View optimization is achieved by continuously and differentiably refining the camera splats so that desirable target distributions are observed from the point cameras, in a manner similar to the original 3D Gaussian splatting. Compared to the Farthest View Sampling (FVS) approach, our optimized views demonstrate superior performance in capturing complex view-dependent phenomena, including intense metallic reflections and intricate textures such as text. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_15677 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Camera Splatting for Continuous View Optimization Lee, Gahye Kim, Hyomin Ju, Gwangjin Son, Jooeun Yoon, Hyejeong Lee, Seungyong Computer Vision and Pattern Recognition We propose Camera Splatting, a novel view optimization framework for novel view synthesis. Each camera is modeled as a 3D Gaussian, referred to as a camera splat, and virtual cameras, termed point cameras, are placed at 3D points sampled near the surface to observe the distribution of camera splats. View optimization is achieved by continuously and differentiably refining the camera splats so that desirable target distributions are observed from the point cameras, in a manner similar to the original 3D Gaussian splatting. Compared to the Farthest View Sampling (FVS) approach, our optimized views demonstrate superior performance in capturing complex view-dependent phenomena, including intense metallic reflections and intricate textures such as text. |
| title | Camera Splatting for Continuous View Optimization |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2509.15677 |