metadata

tags:
  - model_hub_mixin
  - pytorch_model_hub_mixin

This model has been pushed to the Hub using the PytorchModelHubMixin integration:

Library: [More Information Needed]
Docs: [More Information Needed]

license: apache-2.0 - en

LAM: Large Avatar Model for One-shot Animatable Gaussian Head

Overview

This repository contains the example data and face blendshape model of the paper LAM.

LAM creates animatable Gaussian heads with one-shot images in a single forward pass in seconds. The reconstructed Gaussian avatar can be reenacted and rendered on various platforms in real-time.

Quick Start

Please refer to our Github Repo

Download Model

from huggingface_hub import hf_hub_download

hf_hub_download(repo_id="3DAIGC/LAM-assets",
                repo_type='model',
                filename='LAM_human_model.tar',
                local_dir='./')
os.system('tar -xf LAM_human_model.tar && rm LAM_human_model.tar')

# launch example for LAM
hf_hub_download(repo_id='3DAIGC/LAM-assets',
                repo_type='model',
                filename='LAM_assets.tar',
                local_dir='./')
os.system('tar -xf LAM_assets.tar && rm LAM_assets.tar')

Citation

@article{he2025lam,
  title={LAM: Large Avatar Model for One-shot Animatable Gaussian Head},
  author={He, Yisheng and Gu, Xiaodong and Ye, Xiaodan and Xu, Chao and Zhao, Zhengyi and Dong, Yuan and Yuan, Weihao and Dong, Zilong and Bo, Liefeng},
  journal={arXiv preprint arXiv:2502.17796},
  year={2025}
}