BCNet: Learning Body and Cloth Shape from A Single Image

  • 2020-04-01 03:41:36
  • Boyi Jiang, Juyong Zhang, Yang Hong, Jinhao Luo, Ligang Liu, Hujun Bao
  • 28

Abstract

In this paper, we consider the problem to automatically reconstruct bothgarment and body shapes from a single near front view RGB image. To this end,we propose a layered garment representation on top of SMPL and novelly make theskinning weight of garment to be independent with the body mesh, whichsignificantly improves the expression ability of our garment model. Comparedwith existing methods, our method can support more garment categories likeskirts and recover more accurate garment geometry. To train our model, weconstruct two large scale datasets with ground truth body and garmentgeometries as well as paired color images. Compared with single mesh ornon-parametric representation, our method can achieve more flexible controlwith separate meshes, makes applications like re-pose, garment transfer, andgarment texture mapping possible.

 

Quick Read (beta)

loading the full paper ...