vchoutas / smplify-x

Expressive Body Capture: 3D Hands, Face, and Body from a Single Image
https://smpl-x.is.tue.mpg.de/
Other
1.73k stars 335 forks source link

how to run this? #146

Open faheem-khaskheli opened 3 years ago

faheem-khaskheli commented 3 years ago

what is model_folder and vpose_folder when running the code?

!python smplifyx/main.py --config cfg_files/fit_smplx.yaml \ --data_folder DATA_FOLDER \ --output_folder OUTPUT_FOLDER \ --visualize="True" \ --model_folder MODEL_FOLDER \ --vposer_ckpt VPOSER_FOLDER \ --part_segm_fn smplx_parts_segm.pkl

yiweike commented 3 years ago

hey! Did you solve it? I 'm looking for the solution.

swframe commented 3 years ago

Look at the other repo README.md file: https://github.com/vchoutas/smplx You will see the structure of the models folder. You can get the pkl files from the external sites mentioned in the READMEs. At the external sites you will see a download link for vposer.
From that vposer .zip file you should get "vposer_v1_0/snapshots/TR00_E096.pt"

matanox commented 2 years ago

Has anyone gotten that entire stack working with python 3.9 yet? I was struggling with tinkering dependency versions for pip with 3.9 for getting vposer to install per its presumably pre-3.9 requirements.txt.