# DrapeNet: Garment Generation and Self-Supervised Draping

Source: [DrapeNet: Garment Generation and Self-Supervised Draping](<https://www.sota2.com/research/paper/68fa6ecc1e209131cea42364>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-04T05:17:20.802Z

Authors: Luca De Luigi, Ren Li, Beno\\^it Guillard, Mathieu Salzmann, Pascal Fua

Published: 2022-11-21T09:13:53.000Z

## Abstract

Recent approaches to drape garments quickly over arbitrary human bodies leverage self-supervision to eliminate the need for large training sets. However, they are designed to train one network per clothing item, which severely limits their generalization abilities. In our work, we rely on self-supervision to train a single network to drape multiple garments. This is achieved by predicting a 3D deformation field conditioned on the latent codes of a generative network, which models garments as unsigned distance fields. Our pipeline can generate and drape previously unseen garments of any topology, whose shape can be edited by manipulating their latent codes. Being fully differentiable, our formulation makes it possible to recover accurate 3D models of garments from partial observations -- images or 3D scans -- via gradient descent. Our code is publicly available at https://github.com/liren2515/DrapeNet .

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Garment Draping on Unseen garment meshes Top (test)](<https://www.sota2.com/research/sota/garment-draping-on-unseen-garment-meshes-top-test.md>): Garment Draping; Unseen garment meshes Top (test); Strain Energy: 0.43.

- [Garment Draping on Unseen garment meshes Bottom (test)](<https://www.sota2.com/research/sota/garment-draping-on-unseen-garment-meshes-bottom-test.md>): Garment Draping; Unseen garment meshes Bottom (test); Strain Energy: 0.41.

- [3D Garment Reconstruction on Cloth3D (test)](<https://www.sota2.com/research/sota/3d-garment-reconstruction-on-cloth3d-test.md>): 3D Garment Reconstruction; CLOTH3D (test); CD (x10^-4): 0.36.

- [3D Garment Reconstruction on DigitalMe (test)](<https://www.sota2.com/research/sota/3d-garment-reconstruction-on-digitalme-test.md>): 3D Garment Reconstruction; DigitalMe (test); CD (x10^-2): 0.56.

- [Garment Draping on CLOTH3D (test)](<https://www.sota2.com/research/sota/garment-draping-on-cloth3d-test.md>): Garment Draping; CLOTH3D (test); Stretch Error: 0.43.

- [3D Cloth Reconstruction on CustomHumans (test)](<https://www.sota2.com/research/sota/3d-cloth-reconstruction-on-customhumans-test.md>): 3D Cloth Reconstruction; CustomHumans (test); Chamfer Distance: 114.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2211.11277>)

[PDF](<https://arxiv.org/pdf/2211.11277.pdf>)

[Implementation](<https://github.com/liren2515/DrapeNet>)