Neural Light Field 3D Printing


SIGGRAPH Asia 2020 (ACM Transactions on Graphics, 39(6), to appear)

Quan Zheng1     Vahid Babaei1     Gordon Wetzstein2     Hans-Peter Seidel1    
Matthias Zwicker3     Gurprit Singh1

1Max Planck Institute for Informatics    2Stanford University    3University of Maryland, College Park

teaser figure

Our approach encodes input light field imagery (b) as a continuous representation of the volumetric attenuator using neural networks (a). Top row shows the fabricated prototypes using our approach that optimizes both planar (c) and non-planar displays (d). Bottom row compares the central view (red inset) of the input light field (b) with front-view photographs of printed prototypes.

Abstract

Modern 3D printers are capable of printing large-size light-field displays at high-resolutions. However, optimizing such displays in full 3D volume for a given light-field imagery is still a challenging task. Existing light field displays optimize over relatively small resolutions using a few co-planar layers in a 2.5D fashion to keep the problem tractable. In this paper, we propose a novel end-to-end optimization approach that encodes input light field imagery as a continuous-space implicit representation in a neural network. This allows fabricating high-resolution, attenuation-based volumetric displays that exhibit the target light fields. In addition, we incorporate the physical constraints of the material to the optimization such that the result can be printed in practice. Our simulation experiments demonstrate that our approach brings significant visual quality improvement compared to the multilayer and uniform grid-based approaches. We validate our simulations with fabricated prototypes and demonstrate that our pipeline is flexible enough to allow fabrications of both planar and non-planar displays.

Materials

Citation

@article{neurallfprint2020,
title={Neural Light Field 3D Printing},
author={Zheng, Quan and Babaei, Vahid and Wetzstein, Gordon and Seidel, Hans-Peter and Zwicker, Matthias and Singh, Gurprit},
journal={ACM Transactions on Graphics (TOG)},
volume={39},
number={6},
articleno = {207},
year={2020},
numpages = {12},
doi = {10.1145/3414685.3417879},
publisher={ACM New York, NY, USA}
}

Acknowledgments

We would like to thank all anonymous reviewers for their valuable feedback. The first author is funded by the Research Executive Agency 739578.