English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Paper

Ladybird: Quasi-Monte Carlo Sampling for Deep Implicit Field Based 3D Reconstruction with Symmetry

MPS-Authors
/persons/resource/persons256057

Fan,  Tianqi
Computer Graphics, MPI for Informatics, Max Planck Society;

/persons/resource/persons225790

Singh,  Gurprit
Computer Graphics, MPI for Informatics, Max Planck Society;

External Resource
No external resources are shared
Fulltext (public)

arXiv:2007.13393.pdf
(Preprint), 4MB

Supplementary Material (public)
There is no public supplementary material available
Citation

Xu, Y., Fan, T., Yuan, Y., & Singh, G. (2020). Ladybird: Quasi-Monte Carlo Sampling for Deep Implicit Field Based 3D Reconstruction with Symmetry. Retrieved from https://arxiv.org/abs/2007.13393.


Cite as: http://hdl.handle.net/21.11116/0000-0007-CECA-E
Abstract
Deep implicit field regression methods are effective for 3D reconstruction from single-view images. However, the impact of different sampling patterns on the reconstruction quality is not well-understood. In this work, we first study the effect of point set discrepancy on the network training. Based on Farthest Point Sampling algorithm, we propose a sampling scheme that theoretically encourages better generalization performance, and results in fast convergence for SGD-based optimization algorithms. Secondly, based on the reflective symmetry of an object, we propose a feature fusion method that alleviates issues due to self-occlusions which makes it difficult to utilize local image features. Our proposed system Ladybird is able to create high quality 3D object reconstructions from a single input image. We evaluate Ladybird on a large scale 3D dataset (ShapeNet) demonstrating highly competitive results in terms of Chamfer distance, Earth Mover's distance and Intersection Over Union (IoU).