Saved in:
Bibliographic Details
Main Authors: Wenninger, Stephan, Kemper, Fabian, Schwanecke, Ulrich, Botsch, Mario
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2312.02173
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909228452544512
author Wenninger, Stephan
Kemper, Fabian
Schwanecke, Ulrich
Botsch, Mario
author_facet Wenninger, Stephan
Kemper, Fabian
Schwanecke, Ulrich
Botsch, Mario
contents Human shape spaces have been extensively studied, as they are a core element of human shape and pose inference tasks. Classic methods for creating a human shape model register a surface template mesh to a database of 3D scans and use dimensionality reduction techniques, such as Principal Component Analysis, to learn a compact representation. While these shape models enable global shape modifications by correlating anthropometric measurements with the learned subspace, they only provide limited localized shape control. We instead register a volumetric anatomical template, consisting of skeleton bones and soft tissue, to the surface scans of the CAESAR database. We further enlarge our training data to the full Cartesian product of all skeletons and all soft tissues using physically plausible volumetric deformation transfer. This data is then used to learn an anatomically constrained volumetric human shape model in a self-supervised fashion. The resulting TailorMe model enables shape sampling, localized shape manipulation, and fast inference from given surface scans.
format Preprint
id arxiv_https___arxiv_org_abs_2312_02173
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle TailorMe: Self-Supervised Learning of an Anatomically Constrained Volumetric Human Shape Model
Wenninger, Stephan
Kemper, Fabian
Schwanecke, Ulrich
Botsch, Mario
Computer Vision and Pattern Recognition
Graphics
I.3.0; I.5.1
Human shape spaces have been extensively studied, as they are a core element of human shape and pose inference tasks. Classic methods for creating a human shape model register a surface template mesh to a database of 3D scans and use dimensionality reduction techniques, such as Principal Component Analysis, to learn a compact representation. While these shape models enable global shape modifications by correlating anthropometric measurements with the learned subspace, they only provide limited localized shape control. We instead register a volumetric anatomical template, consisting of skeleton bones and soft tissue, to the surface scans of the CAESAR database. We further enlarge our training data to the full Cartesian product of all skeletons and all soft tissues using physically plausible volumetric deformation transfer. This data is then used to learn an anatomically constrained volumetric human shape model in a self-supervised fashion. The resulting TailorMe model enables shape sampling, localized shape manipulation, and fast inference from given surface scans.
title TailorMe: Self-Supervised Learning of an Anatomically Constrained Volumetric Human Shape Model
topic Computer Vision and Pattern Recognition
Graphics
I.3.0; I.5.1
url https://arxiv.org/abs/2312.02173