arXiv Analytics

Sign in

arXiv:1811.10719 [cs.CV]AbstractReferencesReviewsResources

Learning View Priors for Single-view 3D Reconstruction

Hiroharu Kato, Tatsuya Harada

Published 2018-11-26Version 1

There is some ambiguity in the 3D shape of an object when the number of observed views is small. Because of this ambiguity, although a 3D object reconstructor can be trained using a single view or a few views per object, reconstructed shapes only fit the observed views and appear incorrect from the unobserved viewpoints. To reconstruct shapes that look reasonable from any viewpoint, we propose to train a discriminator that learns prior knowledge regarding possible views. The discriminator is trained to distinguish the reconstructed views of the observed viewpoints from those of the unobserved viewpoints. The reconstructor is trained to correct unobserved views by fooling the discriminator. Our method outperforms current state-of-the-art methods on both synthetic and natural image datasets; this validates the effectiveness of our method.

Related articles: Most relevant | Search more
arXiv:2307.12194 [cs.CV] (Published 2023-07-23)
LIST: Learning Implicitly from Spatial Transformers for Single-View 3D Reconstruction
arXiv:2104.14169 [cs.CV] (Published 2021-04-29)
Using Adaptive Gradient for Texture Learning in Single-View 3D Reconstruction
arXiv:2205.06267 [cs.CV] (Published 2022-05-12)
Topologically-Aware Deformation Fields for Single-View 3D Reconstruction