CLOTH3D: Clothed 3D Humans

被引：95
作者：
Bertiche, Hugo [1 ,2 ]
Madadi, Meysam [1 ,2 ]
Escalera, Sergio [1 ,2 ]
机构：
[1] Univ Barcelona, Barcelona, Spain
[2] Comp Vis Ctr, Barcelona, Spain
来源：
COMPUTER VISION - ECCV 2020, PT XX | 2020年 / 12365卷
关键词：
3D; Human; Garment; Cloth; Dataset; Generative model;
D O I：
10.1007/978-3-030-58565-5_21
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
We present CLOTH3D, the first big scale synthetic dataset of 3D clothed human sequences. CLOTH3D contains a large variability on garment type, topology, shape, size, tightness and fabric. Clothes are simulated on top of thousands of different pose sequences and body shapes, generating realistic cloth dynamics. We provide the dataset with a generative model for cloth generation. We propose a Conditional Variational Auto-Encoder (CVAE) based on graph convolutions (GCVAE) to learn garment latent spaces. This allows for realistic generation of 3D garments on top of SMPL model for any pose and shape.
引用
收藏
页码：344 / 359
页数：16
相关论文
共 34 条
[31]   Analyzing Clothing Layer Deformation Statistics of 3D Human Motions [J].
Yang, Jinlong ;
Franco, Jean-Sebastien ;
Hetroy-Wheeler, Franck ;
Wuhrer, Stefanie .
COMPUTER VISION - ECCV 2018, PT VII, 2018, 11211 :245-261
[32]  
Yu T, 2019, Arxiv, DOI arXiv:1903.06323
[33]  
Yuan YJ, 2019, Arxiv, DOI arXiv:1908.02507
[34]   Detailed, accurate, human shape estimation from clothed 3D scan sequences [J].
Zhang, Chao ;
Pujades, Sergi ;
Black, Michael ;
Pons-Moll, Gerard .
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, :5484-5493
1234