Skip to content

Moronrift08/FastAnimate

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Installation

To deploy and run FastAnimate, run the following scripts:

conda create -n fastanimate
pip install -r requirements.txt
conda activate fastanimate

Then, download and compile diff-gaussian-rasterization, kaolin and nvidiffrast.

Download models and data

  • SMPL/SMPL-X model: register and download SMPL and SMPL-X, and put these files in /smpl_related/ to replace the pseudo files. The folder should have the following structure:
smpl_related
 └── models
     └──smpl
       ├── ...
       ├── ...
       └── ...
     └──smplx
       ├── ...
       ├── ...
       └── ...
 └── smpl_data
   ├── ...
   ├── ...
   └── ...

Train

CUDA_VISIBLE_DEVICES=0,1,2,3 nohup accelerate launch --config_file acc_configs/gpu4.yaml train.py  big --workspace workspace_train --resume ./path_to_your_stage_1_models

Test

CUDA_VISIBLE_DEVICES=0 python test.py big --resume ./path_to_your_stage_2_models --workspace workspace_test

Acknowledgements

This project is built on source codes shared by LGM, SiTH, MVDream and MultiGo.

About

The official repo for "FastAnimate: Towards Learnable Template Construction and Pose Deformation for Fast 3D Human Avatar Animation"

Resources

License

Stars

6 stars

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors