Monocular Dynamic Gaussian Splatting: Fast, Brittle, and Scene Complexity Rules
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916782575452160 |
|---|---|
| author | Liang, Yiqing Okunev, Mikhail Uy, Mikaela Angelina Li, Runfeng Guibas, Leonidas Tompkin, James Harley, Adam W. |
| author_facet | Liang, Yiqing Okunev, Mikhail Uy, Mikaela Angelina Li, Runfeng Guibas, Leonidas Tompkin, James Harley, Adam W. |
| contents | Gaussian splatting methods are emerging as a popular approach for converting multi-view image data into scene representations that allow view synthesis. In particular, there is interest in enabling view synthesis for dynamic scenes using only monocular input data -- an ill-posed and challenging problem. The fast pace of work in this area has produced multiple simultaneous papers that claim to work best, which cannot all be true. In this work, we organize, benchmark, and analyze many Gaussian-splatting-based methods, providing apples-to-apples comparisons that prior works have lacked. We use multiple existing datasets and a new instructive synthetic dataset designed to isolate factors that affect reconstruction quality. We systematically categorize Gaussian splatting methods into specific motion representation types and quantify how their differences impact performance. Empirically, we find that their rank order is well-defined in synthetic data, but the complexity of real-world data currently overwhelms the differences. Furthermore, the fast rendering speed of all Gaussian-based methods comes at the cost of brittleness in optimization. We summarize our experiments into a list of findings that can help to further progress in this lively problem setting. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_04457 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Monocular Dynamic Gaussian Splatting: Fast, Brittle, and Scene Complexity Rules Liang, Yiqing Okunev, Mikhail Uy, Mikaela Angelina Li, Runfeng Guibas, Leonidas Tompkin, James Harley, Adam W. Computer Vision and Pattern Recognition Gaussian splatting methods are emerging as a popular approach for converting multi-view image data into scene representations that allow view synthesis. In particular, there is interest in enabling view synthesis for dynamic scenes using only monocular input data -- an ill-posed and challenging problem. The fast pace of work in this area has produced multiple simultaneous papers that claim to work best, which cannot all be true. In this work, we organize, benchmark, and analyze many Gaussian-splatting-based methods, providing apples-to-apples comparisons that prior works have lacked. We use multiple existing datasets and a new instructive synthetic dataset designed to isolate factors that affect reconstruction quality. We systematically categorize Gaussian splatting methods into specific motion representation types and quantify how their differences impact performance. Empirically, we find that their rank order is well-defined in synthetic data, but the complexity of real-world data currently overwhelms the differences. Furthermore, the fast rendering speed of all Gaussian-based methods comes at the cost of brittleness in optimization. We summarize our experiments into a list of findings that can help to further progress in this lively problem setting. |
| title | Monocular Dynamic Gaussian Splatting: Fast, Brittle, and Scene Complexity Rules |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2412.04457 |