MVDiffusion++: A Dense High-resolution Multi-view Diffusion Model for Single or Sparse-view 3D Object Reconstruction
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_ | 1866910428000419840 |
|---|---|
| author | Tang, Shitao Chen, Jiacheng Wang, Dilin Tang, Chengzhou Zhang, Fuyang Fan, Yuchen Chandra, Vikas Furukawa, Yasutaka Ranjan, Rakesh |
| author_facet | Tang, Shitao Chen, Jiacheng Wang, Dilin Tang, Chengzhou Zhang, Fuyang Fan, Yuchen Chandra, Vikas Furukawa, Yasutaka Ranjan, Rakesh |
| contents | This paper presents a neural architecture MVDiffusion++ for 3D object reconstruction that synthesizes dense and high-resolution views of an object given one or a few images without camera poses. MVDiffusion++ achieves superior flexibility and scalability with two surprisingly simple ideas: 1) A ``pose-free architecture'' where standard self-attention among 2D latent features learns 3D consistency across an arbitrary number of conditional and generation views without explicitly using camera pose information; and 2) A ``view dropout strategy'' that discards a substantial number of output views during training, which reduces the training-time memory footprint and enables dense and high-resolution view synthesis at test time. We use the Objaverse for training and the Google Scanned Objects for evaluation with standard novel view synthesis and 3D reconstruction metrics, where MVDiffusion++ significantly outperforms the current state of the arts. We also demonstrate a text-to-3D application example by combining MVDiffusion++ with a text-to-image generative model. The project page is at https://mvdiffusion-plusplus.github.io. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_12712 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MVDiffusion++: A Dense High-resolution Multi-view Diffusion Model for Single or Sparse-view 3D Object Reconstruction Tang, Shitao Chen, Jiacheng Wang, Dilin Tang, Chengzhou Zhang, Fuyang Fan, Yuchen Chandra, Vikas Furukawa, Yasutaka Ranjan, Rakesh Computer Vision and Pattern Recognition This paper presents a neural architecture MVDiffusion++ for 3D object reconstruction that synthesizes dense and high-resolution views of an object given one or a few images without camera poses. MVDiffusion++ achieves superior flexibility and scalability with two surprisingly simple ideas: 1) A ``pose-free architecture'' where standard self-attention among 2D latent features learns 3D consistency across an arbitrary number of conditional and generation views without explicitly using camera pose information; and 2) A ``view dropout strategy'' that discards a substantial number of output views during training, which reduces the training-time memory footprint and enables dense and high-resolution view synthesis at test time. We use the Objaverse for training and the Google Scanned Objects for evaluation with standard novel view synthesis and 3D reconstruction metrics, where MVDiffusion++ significantly outperforms the current state of the arts. We also demonstrate a text-to-3D application example by combining MVDiffusion++ with a text-to-image generative model. The project page is at https://mvdiffusion-plusplus.github.io. |
| title | MVDiffusion++: A Dense High-resolution Multi-view Diffusion Model for Single or Sparse-view 3D Object Reconstruction |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2402.12712 |