Fitting SMPL

[kfarivar]

Hi

I have tried running the :

python smpl_registration/fit_SMPL_IPNet.py <scan_path.obj> <scan_labels.npy> <scale_file.npy> <save_path>

but I get the following error:

Traceback (most recent call last):
  File "smpl_registration/fit_SMPL_IPNet.py", line 381, in <module>
    _, _, _ = fit_SMPL([args.inner_path], scan_labels=[args.inner_labels], display=args.display, save_path=args.save_path,
  File "smpl_registration/fit_SMPL_IPNet.py", line 212, in fit_SMPL
    smpl_faces = sp.get_faces()
  File "/home/logi/Desktop/ipnet/IPNet/lib/smpl_paths.py", line 111, in get_faces
    return np.load(fname)
  File "/home/logi/anaconda3/envs/ipnet/lib/python3.8/site-packages/numpy/lib/npyio.py", line 416, in load
    fid = stack.enter_context(open(os_fspath(file), "rb"))
FileNotFoundError: [Errno 2] No such file or directory: '/BS/RVH/work/data/template/faces.npy'

So my question is:

Do I need to download the smpl models from the paper's website ? if yes where should I put them ?

Also what should the faces.npy file look like ? and how can I get it ?

[kfarivar]

I downloaded the smpl models from their website (v1.1) and put them in: IPNet/smplpytorch/smplpytorch/native/models/. you also need to rename the models to male_model.pkl and female_model.pkl

So for the faces I just made an SMPL model and got the faces from it like this (in the smpl_paths.py):

@staticmethod
    def get_faces():
        from lib.smpl_layer import SMPL_Layer
        model = SMPL_Layer(center_idx=0, gender='male', model_root='smplpytorch/smplpytorch/native/models')
        return model.th_faces.numpy()

I needed to change many absolute paths that were hardcoded to the correct ones to make it work. I suggest running the fitting script and correcting them according to the errors you get.