# Learning to Dress 3D People in Generative Clothing

## TL;DR

This paper introduces CAPE, a generative 3D clothing model that extends SMPL to realistically dress 3D human meshes across various poses and styles, capturing detailed wrinkle and shape variations.

## Contribution

We develop the first generative model that directly dresses 3D human meshes with clothing, conditioned on pose and style, improving realism and generalization over prior minimally-clothed models.

## Key findings

- Successfully models pose-dependent clothing deformation
- Generates diverse clothing styles for different body shapes
- Preserves wrinkle detail in 3D clothing generation

## Abstract

Three-dimensional 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 patchwise 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 at https://cape.is.tue.mpg.de.