3D Reconstruction By Parameterized Surface Mapping

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

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.
BibTeX
@inproceedings{langlois2021d,
  author = {Pierre-Alain Langlois and Matthew Fisher and Oliver Wang and Vladimir G. Kim and Alexandre Boulch and Renaud Marlet and Bryan C. Russell},
  title = {{3D Reconstruction By Parameterized Surface Mapping}},
  booktitle = {ICIP},
  year = {2021},
  doi = {10.1109/icip42928.2021.9506425},
}

Loading related content...

Figure