LHM

🚀 LHM: Large Animatable Human Reconstruction Model for Single Image to 3D in Seconds

LHM is a feed - forward model that can achieve animatable 3D human reconstruction from a single image in seconds, showing strong generalization ability in diverse real - world scenarios.

🚀 Quick Start

Please refer to our Github Repo.

Download Model

from huggingface_hub import snapshot_download 
# 500M-HF Model
model_dir = snapshot_download(repo_id='3DAIGC/LHM-500M-HF', cache_dir='./pretrained_models/huggingface')
# 500M Model
model_dir = snapshot_download(repo_id='3DAIGC/LHM-500M', cache_dir='./pretrained_models/huggingface')
# 1B Model
model_dir = snapshot_download(repo_id='3DAIGC/LHM-1B', cache_dir='./pretrained_models/huggingface')

✨ Features

This repository contains the models of the paper LHM: Large Animatable Human Reconstruction Model for Single Image to 3D in Seconds. LHM is a feed - forward model for animatable 3D human reconstruction from a single image in seconds. Trained on a large - scale video dataset with an image reconstruction loss, our model exhibits strong generalization ability to diverse real - world scenarios.

📄 License

This project is licensed under the Apache 2.0 License.

📚 Documentation

Citation

@inproceedings{qiu2025LHM,
  title={LHM: Large Animatable Human Reconstruction Model from a Single Image in Seconds},
  author={Lingteng Qiu and Xiaodong Gu and Peihao Li  and Qi Zuo
     and Weichao Shen and Junfei Zhang and Kejie Qiu and Weihao Yuan
     and Guanying Chen and Zilong Dong and Liefeng Bo 
    },
  booktitle={arXiv preprint arXiv:2503.10625},
  year={2025}
}