3D Reconstruction by Parameterized Surface Mapping

Pierre-Alain Langlois  Matthew Fisher  Oliver Wang  Vladimir Kim  Alexandre Boulch  Renaud Marlet  Bryan Russell

ICIP 2021

Paper  

Abstract

We introduce an approach for computing a 3D mesh from one or more views of an object by establishing dense correspondences between pixels in the views and 3D locations on a learnable parameterized surface. We propose a multi-view shape encoder that can be jointly trained with the AtlasNet surface parameterization. The shape is further refined using a novel geometric cycle-consistency loss between the learnable parameterized surface and input views. We demonstrate the efficacy of our approach on the ShapeNet-COCO dataset. </a>


BibTeX

@inproceedings{langlois20213d,
  title={3D Reconstruction By Parameterized Surface Mapping},
  author={Langlois, Pierre-Alain and Fisher, Matthew and Wang, Oliver and Kim, Vladimir and Boulch, Alexandre and Marlet, Renaud and Russell, Bryan},
  booktitle={2021 IEEE International Conference on Image Processing (ICIP)},
  pages={3273--3277},
  year={2021},
  organization={IEEE}
}