Repurposing 2D Diffusion Models with Gaussian Atlas for 3D Generation
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912524201361408 |
|---|---|
| author | Xiang, Tiange Li, Kai Long, Chengjiang Häne, Christian Guo, Peihong Delp, Scott Adeli, Ehsan Fei-Fei, Li |
| author_facet | Xiang, Tiange Li, Kai Long, Chengjiang Häne, Christian Guo, Peihong Delp, Scott Adeli, Ehsan Fei-Fei, Li |
| contents | Recent advances in text-to-image diffusion models have been driven by the increasing availability of paired 2D data. However, the development of 3D diffusion models has been hindered by the scarcity of high-quality 3D data, resulting in less competitive performance compared to their 2D counterparts. To address this challenge, we propose repurposing pre-trained 2D diffusion models for 3D object generation. We introduce Gaussian Atlas, a novel representation that utilizes dense 2D grids, enabling the fine-tuning of 2D diffusion models to generate 3D Gaussians. Our approach demonstrates successful transfer learning from a pre-trained 2D diffusion model to a 2D manifold flattened from 3D structures. To support model training, we compile GaussianVerse, a large-scale dataset comprising 205K high-quality 3D Gaussian fittings of various 3D objects. Our experimental results show that text-to-image diffusion models can be effectively adapted for 3D content generation, bridging the gap between 2D and 3D modeling. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_15877 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Repurposing 2D Diffusion Models with Gaussian Atlas for 3D Generation Xiang, Tiange Li, Kai Long, Chengjiang Häne, Christian Guo, Peihong Delp, Scott Adeli, Ehsan Fei-Fei, Li Computer Vision and Pattern Recognition Recent advances in text-to-image diffusion models have been driven by the increasing availability of paired 2D data. However, the development of 3D diffusion models has been hindered by the scarcity of high-quality 3D data, resulting in less competitive performance compared to their 2D counterparts. To address this challenge, we propose repurposing pre-trained 2D diffusion models for 3D object generation. We introduce Gaussian Atlas, a novel representation that utilizes dense 2D grids, enabling the fine-tuning of 2D diffusion models to generate 3D Gaussians. Our approach demonstrates successful transfer learning from a pre-trained 2D diffusion model to a 2D manifold flattened from 3D structures. To support model training, we compile GaussianVerse, a large-scale dataset comprising 205K high-quality 3D Gaussian fittings of various 3D objects. Our experimental results show that text-to-image diffusion models can be effectively adapted for 3D content generation, bridging the gap between 2D and 3D modeling. |
| title | Repurposing 2D Diffusion Models with Gaussian Atlas for 3D Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2503.15877 |