GRM: Large Gaussian Reconstruction Model for Efficient 3D Reconstruction and Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xu, Yinghao, Shi, Zifan, Yifan, Wang, Chen, Hansheng, Yang, Ceyuan, Peng, Sida, Shen, Yujun, Wetzstein, Gordon
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929284979884032
author Xu, Yinghao
Shi, Zifan
Yifan, Wang
Chen, Hansheng
Yang, Ceyuan
Peng, Sida
Shen, Yujun
Wetzstein, Gordon
author_facet Xu, Yinghao
Shi, Zifan
Yifan, Wang
Chen, Hansheng
Yang, Ceyuan
Peng, Sida
Shen, Yujun
Wetzstein, Gordon
contents We introduce GRM, a large-scale reconstructor capable of recovering a 3D asset from sparse-view images in around 0.1s. GRM is a feed-forward transformer-based model that efficiently incorporates multi-view information to translate the input pixels into pixel-aligned Gaussians, which are unprojected to create a set of densely distributed 3D Gaussians representing a scene. Together, our transformer architecture and the use of 3D Gaussians unlock a scalable and efficient reconstruction framework. Extensive experimental results demonstrate the superiority of our method over alternatives regarding both reconstruction quality and efficiency. We also showcase the potential of GRM in generative tasks, i.e., text-to-3D and image-to-3D, by integrating it with existing multi-view diffusion models. Our project website is at: https://justimyhxu.github.io/projects/grm/.
format Preprint
id arxiv_https___arxiv_org_abs_2403_14621
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle GRM: Large Gaussian Reconstruction Model for Efficient 3D Reconstruction and Generation
Xu, Yinghao
Shi, Zifan
Yifan, Wang
Chen, Hansheng
Yang, Ceyuan
Peng, Sida
Shen, Yujun
Wetzstein, Gordon
Computer Vision and Pattern Recognition
We introduce GRM, a large-scale reconstructor capable of recovering a 3D asset from sparse-view images in around 0.1s. GRM is a feed-forward transformer-based model that efficiently incorporates multi-view information to translate the input pixels into pixel-aligned Gaussians, which are unprojected to create a set of densely distributed 3D Gaussians representing a scene. Together, our transformer architecture and the use of 3D Gaussians unlock a scalable and efficient reconstruction framework. Extensive experimental results demonstrate the superiority of our method over alternatives regarding both reconstruction quality and efficiency. We also showcase the potential of GRM in generative tasks, i.e., text-to-3D and image-to-3D, by integrating it with existing multi-view diffusion models. Our project website is at: https://justimyhxu.github.io/projects/grm/.
title GRM: Large Gaussian Reconstruction Model for Efficient 3D Reconstruction and Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2403.14621