ClotheDreamer: Text-Guided Garment Generation with 3D Gaussians

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liu, Yufei, Tang, Junshu, Zheng, Chu, Zhang, Shijie, Hao, Jinkun, Zhu, Junwei, Huang, Dongjin
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914847480872960
author Liu, Yufei
Tang, Junshu
Zheng, Chu
Zhang, Shijie
Hao, Jinkun
Zhu, Junwei
Huang, Dongjin
author_facet Liu, Yufei
Tang, Junshu
Zheng, Chu
Zhang, Shijie
Hao, Jinkun
Zhu, Junwei
Huang, Dongjin
contents High-fidelity 3D garment synthesis from text is desirable yet challenging for digital avatar creation. Recent diffusion-based approaches via Score Distillation Sampling (SDS) have enabled new possibilities but either intricately couple with human body or struggle to reuse. We introduce ClotheDreamer, a 3D Gaussian-based method for generating wearable, production-ready 3D garment assets from text prompts. We propose a novel representation Disentangled Clothe Gaussian Splatting (DCGS) to enable separate optimization. DCGS represents clothed avatar as one Gaussian model but freezes body Gaussian splats. To enhance quality and completeness, we incorporate bidirectional SDS to supervise clothed avatar and garment RGBD renderings respectively with pose conditions and propose a new pruning strategy for loose clothing. Our approach can also support custom clothing templates as input. Benefiting from our design, the synthetic 3D garment can be easily applied to virtual try-on and support physically accurate animation. Extensive experiments showcase our method's superior and competitive performance. Our project page is at https://ggxxii.github.io/clothedreamer.
format Preprint
id arxiv_https___arxiv_org_abs_2406_16815
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle ClotheDreamer: Text-Guided Garment Generation with 3D Gaussians
Liu, Yufei
Tang, Junshu
Zheng, Chu
Zhang, Shijie
Hao, Jinkun
Zhu, Junwei
Huang, Dongjin
Computer Vision and Pattern Recognition
High-fidelity 3D garment synthesis from text is desirable yet challenging for digital avatar creation. Recent diffusion-based approaches via Score Distillation Sampling (SDS) have enabled new possibilities but either intricately couple with human body or struggle to reuse. We introduce ClotheDreamer, a 3D Gaussian-based method for generating wearable, production-ready 3D garment assets from text prompts. We propose a novel representation Disentangled Clothe Gaussian Splatting (DCGS) to enable separate optimization. DCGS represents clothed avatar as one Gaussian model but freezes body Gaussian splats. To enhance quality and completeness, we incorporate bidirectional SDS to supervise clothed avatar and garment RGBD renderings respectively with pose conditions and propose a new pruning strategy for loose clothing. Our approach can also support custom clothing templates as input. Benefiting from our design, the synthetic 3D garment can be easily applied to virtual try-on and support physically accurate animation. Extensive experiments showcase our method's superior and competitive performance. Our project page is at https://ggxxii.github.io/clothedreamer.
title ClotheDreamer: Text-Guided Garment Generation with 3D Gaussians
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2406.16815