MeshLRM: Large Reconstruction Model for High-Quality Meshes
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917899775508480 |
|---|---|
| author | Wei, Xinyue Zhang, Kai Bi, Sai Tan, Hao Luan, Fujun Deschaintre, Valentin Sunkavalli, Kalyan Su, Hao Xu, Zexiang |
| author_facet | Wei, Xinyue Zhang, Kai Bi, Sai Tan, Hao Luan, Fujun Deschaintre, Valentin Sunkavalli, Kalyan Su, Hao Xu, Zexiang |
| contents | We propose MeshLRM, a novel LRM-based approach that can reconstruct a high-quality mesh from merely four input images in less than one second. Different from previous large reconstruction models (LRMs) that focus on NeRF-based reconstruction, MeshLRM incorporates differentiable mesh extraction and rendering within the LRM framework. This allows for end-to-end mesh reconstruction by fine-tuning a pre-trained NeRF LRM with mesh rendering. Moreover, we improve the LRM architecture by simplifying several complex designs in previous LRMs. MeshLRM's NeRF initialization is sequentially trained with low- and high-resolution images; this new LRM training strategy enables significantly faster convergence and thereby leads to better quality with less compute. Our approach achieves state-of-the-art mesh reconstruction from sparse-view inputs and also allows for many downstream applications, including text-to-3D and single-image-to-3D generation. Project page: https://sarahweiii.github.io/meshlrm/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_12385 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MeshLRM: Large Reconstruction Model for High-Quality Meshes Wei, Xinyue Zhang, Kai Bi, Sai Tan, Hao Luan, Fujun Deschaintre, Valentin Sunkavalli, Kalyan Su, Hao Xu, Zexiang Computer Vision and Pattern Recognition Graphics We propose MeshLRM, a novel LRM-based approach that can reconstruct a high-quality mesh from merely four input images in less than one second. Different from previous large reconstruction models (LRMs) that focus on NeRF-based reconstruction, MeshLRM incorporates differentiable mesh extraction and rendering within the LRM framework. This allows for end-to-end mesh reconstruction by fine-tuning a pre-trained NeRF LRM with mesh rendering. Moreover, we improve the LRM architecture by simplifying several complex designs in previous LRMs. MeshLRM's NeRF initialization is sequentially trained with low- and high-resolution images; this new LRM training strategy enables significantly faster convergence and thereby leads to better quality with less compute. Our approach achieves state-of-the-art mesh reconstruction from sparse-view inputs and also allows for many downstream applications, including text-to-3D and single-image-to-3D generation. Project page: https://sarahweiii.github.io/meshlrm/ |
| title | MeshLRM: Large Reconstruction Model for High-Quality Meshes |
| topic | Computer Vision and Pattern Recognition Graphics |
| url | https://arxiv.org/abs/2404.12385 |