Hi!
You have provided the pre-trained model. We would love to use it for our work but you have not provided with an evaluation script that we could use. I do see it being used within the pose_annotator. But, it is difficult to find out whats necessary for standalone execution. If you do have eval scripts, please do upload them. Kindly provide the same. Further, pytorch3d does not have a supported build for python 3.7. How did you manage to install pytorch3d in your python 3.7 environment (MANO repo you use also requires 3.7) - refer to https://github.com/facebookresearch/pytorch3d/releases/tag/v0.7.2 for reference for pytorch3d's declined support for python 3.7.
Thanks for the great work!