conference paper
UCLID-Net: Single View Reconstruction in Object Space
2020
Advances in Neural Information Processing Systems 33
Most state-of-the-art deep geometric learning single-view reconstruction approaches rely on encoder-decoder architectures that output either shape parametrizations or implicit representations. However, these representations rarely preserve the Euclidean structure of the 3D space objects exist in. In this paper, we show that building a geometry preserving 3-dimensional latent space helps the network concurrently learn global shape regularities and local reasoning in the object coordinate space and, as a result, boosts performance.
Type
conference paper
Author(s)
Date Issued
2020
Published in
Advances in Neural Information Processing Systems 33
Total of pages
8
Subjects
Editorial or Peer reviewed
REVIEWED
Written at
EPFL
EPFL units
Event name | Event place | Event date |
Virtual | December 6-12, 2020 | |
Available on Infoscience
October 28, 2020
Use this identifier to reference this record