Learning to Dress 3D People in Generative Clothing (CVPR 2020)

18 August 2020

04:05

Abstract: 3D human body models are widely used in the analysis of human pose and motion. Existing models, however, are learned from minimally-clothed 3D scans and thus do not generalize to the complexity of dressed people in common images and videos. Additionally, current models lack the expressive power needed to represent the complex non-linear geometry of pose-dependent clothing shapes. To address this, we learn a generative 3D mesh model of clothed people from 3D scans with varying pose and clothing. Specifically, we train a conditional Mesh-VAE-GAN to learn the clothing deformation from the SMPL body model, making clothing an additional term in SMPL. Our model is conditioned on both pose and clothing type, giving the ability to draw samples of clothing to dress different body shapes in a variety of styles and poses. To preserve wrinkle detail, our Mesh-VAE-GAN extends patch-wise discriminators to 3D meshes. Our model, named CAPE, represents global shape and fine local structure, effectively extending the SMPL body model to clothing. To our knowledge, this is the first generative model that directly dresses 3D human body meshes and generalizes to different poses. The model, code and data are available for research purposes. Code: https://github.com/QianliM/CAPE Dataset: https://cape.is.tue.mpg.de/ Paper: https://arxiv.org/abs/1907.13615 Citation: @inproceedings{CAPE:CVPR:20, title = {Learning to Dress {3D} People in Generative Clothing}, author = {Ma, Qianli and Yang, Jinlong and Ranjan, Anurag and Pujades, Sergi and Pons-Moll, Gerard and Tang, Siyu and Black, Michael J.}, booktitle = {Computer Vision and Pattern Recognition (CVPR)}, month = June, year = {2020}, pages={6468--6477}, month_numeric = {6}}