VolumetricSMPL

Description

VolumetricSMPL is an extension of the SMPL body model that incorporates a volumetric (signed distance field, SDF) representation. This enables seamless interaction with 3D geometries, such as scenes, objects, and other humans.

Installation

Ensure that PyTorch and PyTorch3D are installed with GPU support. Then, install VolumetricSMPL via:

pip install VolumetricSMPL

Usage

VolumetricSMPL extends the interface of the SMPL-X package by attaching a volumetric representation to the body model. This allows for querying signed distance fields for arbitrary points and accessing collision loss terms.

A more detailed tutorial is available here, demonstrating how to integrate VolumetricSMPL into applications requiring human-scene, human-object, and human-human interactions.

Example Usage

import smplx
from VolumetricSMPL import attach_volume

# Create a SMPL body and extend it with volumetric functionalities (supports SMPL, SMPLH, and SMPL-X)
model = smplx.create(**smpl_parameters)
attach_volume(model)

# Forward pass
smpl_output = model(**smpl_data)  

# Ensure valid SMPL variables (pose parameters, joints, and vertices)
assert model.joint_mapper is None, "VolumetricSMPL requires valid SMPL joints as input."

# Access volumetric functionalities
model.volume.query(scan_point_cloud)                 # Query SDF for given points
model.volume.selfpen_loss(smpl_output)               # Compute self-intersection loss
model.volume.collision_loss(smpl_output, scan_point_cloud)  # Compute collisions with external geometries

Pretrained Models

Pretrained models are automatically fetched and loaded. They can also be found in the dev branch inside the ./models directory.

Contact

For questions, please contact Marko Mihajlovic (markomih@ethz.ch) or open an issue on GitHub.