Enhancing Monocular 3D Scene Completion with Diffusion Model
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915178843471872 |
|---|---|
| author | Song, Changlin Wang, Jiaqi Zhu, Liyun Weng, He |
| author_facet | Song, Changlin Wang, Jiaqi Zhu, Liyun Weng, He |
| contents | 3D scene reconstruction is essential for applications in virtual reality, robotics, and autonomous driving, enabling machines to understand and interact with complex environments. Traditional 3D Gaussian Splatting techniques rely on images captured from multiple viewpoints to achieve optimal performance, but this dependence limits their use in scenarios where only a single image is available. In this work, we introduce FlashDreamer, a novel approach for reconstructing a complete 3D scene from a single image, significantly reducing the need for multi-view inputs. Our approach leverages a pre-trained vision-language model to generate descriptive prompts for the scene, guiding a diffusion model to produce images from various perspectives, which are then fused to form a cohesive 3D reconstruction. Extensive experiments show that our method effectively and robustly expands single-image inputs into a comprehensive 3D scene, extending monocular 3D reconstruction capabilities without further training. Our code is available https://github.com/CharlieSong1999/FlashDreamer/tree/main. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_00726 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Enhancing Monocular 3D Scene Completion with Diffusion Model Song, Changlin Wang, Jiaqi Zhu, Liyun Weng, He Graphics Artificial Intelligence Computer Vision and Pattern Recognition 3D scene reconstruction is essential for applications in virtual reality, robotics, and autonomous driving, enabling machines to understand and interact with complex environments. Traditional 3D Gaussian Splatting techniques rely on images captured from multiple viewpoints to achieve optimal performance, but this dependence limits their use in scenarios where only a single image is available. In this work, we introduce FlashDreamer, a novel approach for reconstructing a complete 3D scene from a single image, significantly reducing the need for multi-view inputs. Our approach leverages a pre-trained vision-language model to generate descriptive prompts for the scene, guiding a diffusion model to produce images from various perspectives, which are then fused to form a cohesive 3D reconstruction. Extensive experiments show that our method effectively and robustly expands single-image inputs into a comprehensive 3D scene, extending monocular 3D reconstruction capabilities without further training. Our code is available https://github.com/CharlieSong1999/FlashDreamer/tree/main. |
| title | Enhancing Monocular 3D Scene Completion with Diffusion Model |
| topic | Graphics Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2503.00726 |