Published December 23, 2021 | Version 1.0
Dataset Open

AUTH-OpenDR SMPL+D Human Bodies Dataset

  • 1. Aristotle University of Thessaloniki

Description

The  dataset contains 2982 human bodies in various shapes and textures folowing the SMPL-D template. At its core, our dataset  consists of 183 unique SMPL+D bodies, which were generated through non-rigid shape registration of manually generated MakeHuman models. The rest were generated by applying shape and texture alterations to those models. In addition, we provide code for converting those human models in the FBX format.  However, pose-dependent deformations are not applied to the human models. Finally, instructions for setting a demo project in the Webots simulator are provided. In the project, one the SMPL+D models in FBX format can perform an animation from AMASS.

The dataset was developed by Aristotle University of Thessaloniki (AUTH) within the H2020 OpenDR Project.

The dataset is available through the official GitHub repository of the OpenDR toolkit here.

Notes

The dataset, consists of (1) the SMPL+D human bodies in raw format (texture-maps, displacement-maps, shape parameters, gender), (2) the SMPL+D human models in FBX files, (3) a Webots project, (4) code for processing (1), (2), (3). The dataset is licensed under the Creative Commons Attribution 4.0 International License.

Files

SMPLD_human_bodies.pdf

Files (265.3 kB)

Name Size Download all
md5:c5b24f945a8467639d5a27cb0b3cdb98
265.3 kB Preview Download

Additional details

Funding

European Commission
OpenDR - Open Deep Learning Toolkit for Robotics 871449