LGM / README.md
JCTN's picture
Update README.md
2596617 verified
metadata
license: mit
pipeline_tag: text-to-3d
tags:
  - image-to-3d
  - text-to-3d

LGM

This model contains the pretrained weights for LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation.

Project Page | Arxiv | Weights

Introduction

LGM can generate 3D objects from image or text within 5 seconds at high-resolution based on Gaussian Splatting.

Model Details

The model is trained on a ~80K subset of Objaverse. For more details, please refer to our paper.

Usage

To download the model:

from huggingface_hub import hf_hub_download
ckpt_path = hf_hub_download(repo_id="ashawkey/LGM", filename="model_fp16.safetensors")

Please refer to our repo for more details on loading and inference.

Citation

@article{tang2024lgm,
  title={LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation},
  author={Tang, Jiaxiang and Chen, Zhaoxi and Chen, Xiaokang and Wang, Tengfei and Zeng, Gang and Liu, Ziwei},
  journal={arXiv preprint arXiv:2402.05054},
  year={2024}
}