Experimental digital Gabor hologram rendering by a model-trained convolutional neural network
Image and Video Processing
2020-04-21 v1 Data Analysis, Statistics and Probability
Optics
Abstract
Digital hologram rendering can be performed by a convolutional neural network, trained with image pairs calculated by numerical wave propagation from sparse generating images. 512-by-512 pixeldigital Gabor magnitude holograms are successfully estimated from experimental interferograms by a standard UNet trained with 50,000 synthetic image pairs over 70 epochs.
Cite
@article{arxiv.2004.09126,
title = {Experimental digital Gabor hologram rendering by a model-trained convolutional neural network},
author = {J. Rivet and A. Taliercio and C. Fang and G. Tochon and T. Géraud and JP. Huignard and M. Atlan},
journal= {arXiv preprint arXiv:2004.09126},
year = {2020}
}