Generalizable Human Gaussians for Sparse View Synthesis
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_ | 1866913434540441600 |
|---|---|
| author | Kwon, Youngjoong Fang, Baole Lu, Yixing Dong, Haoye Zhang, Cheng Carrasco, Francisco Vicente Mosella-Montoro, Albert Xu, Jianjin Takagi, Shingo Kim, Daeil Prakash, Aayush De la Torre, Fernando |
| author_facet | Kwon, Youngjoong Fang, Baole Lu, Yixing Dong, Haoye Zhang, Cheng Carrasco, Francisco Vicente Mosella-Montoro, Albert Xu, Jianjin Takagi, Shingo Kim, Daeil Prakash, Aayush De la Torre, Fernando |
| contents | Recent progress in neural rendering has brought forth pioneering methods, such as NeRF and Gaussian Splatting, which revolutionize view rendering across various domains like AR/VR, gaming, and content creation. While these methods excel at interpolating {\em within the training data}, the challenge of generalizing to new scenes and objects from very sparse views persists. Specifically, modeling 3D humans from sparse views presents formidable hurdles due to the inherent complexity of human geometry, resulting in inaccurate reconstructions of geometry and textures. To tackle this challenge, this paper leverages recent advancements in Gaussian Splatting and introduces a new method to learn generalizable human Gaussians that allows photorealistic and accurate view-rendering of a new human subject from a limited set of sparse views in a feed-forward manner. A pivotal innovation of our approach involves reformulating the learning of 3D Gaussian parameters into a regression process defined on the 2D UV space of a human template, which allows leveraging the strong geometry prior and the advantages of 2D convolutions. In addition, a multi-scaffold is proposed to effectively represent the offset details. Our method outperforms recent methods on both within-dataset generalization as well as cross-dataset generalization settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_12777 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Generalizable Human Gaussians for Sparse View Synthesis Kwon, Youngjoong Fang, Baole Lu, Yixing Dong, Haoye Zhang, Cheng Carrasco, Francisco Vicente Mosella-Montoro, Albert Xu, Jianjin Takagi, Shingo Kim, Daeil Prakash, Aayush De la Torre, Fernando Computer Vision and Pattern Recognition Graphics Recent progress in neural rendering has brought forth pioneering methods, such as NeRF and Gaussian Splatting, which revolutionize view rendering across various domains like AR/VR, gaming, and content creation. While these methods excel at interpolating {\em within the training data}, the challenge of generalizing to new scenes and objects from very sparse views persists. Specifically, modeling 3D humans from sparse views presents formidable hurdles due to the inherent complexity of human geometry, resulting in inaccurate reconstructions of geometry and textures. To tackle this challenge, this paper leverages recent advancements in Gaussian Splatting and introduces a new method to learn generalizable human Gaussians that allows photorealistic and accurate view-rendering of a new human subject from a limited set of sparse views in a feed-forward manner. A pivotal innovation of our approach involves reformulating the learning of 3D Gaussian parameters into a regression process defined on the 2D UV space of a human template, which allows leveraging the strong geometry prior and the advantages of 2D convolutions. In addition, a multi-scaffold is proposed to effectively represent the offset details. Our method outperforms recent methods on both within-dataset generalization as well as cross-dataset generalization settings. |
| title | Generalizable Human Gaussians for Sparse View Synthesis |
| topic | Computer Vision and Pattern Recognition Graphics |
| url | https://arxiv.org/abs/2407.12777 |